Total Synthesis OfNatural Products, Volume1 Edited by John Apsimon Copyright © 1973 by John Wiley & Sons, Inc.
THE TOTAL SYNTHESIS OF NATURAL PRODUCTS
Total Synthesis OfNatural Products, Volume1 Edited by John Apsimon Copyright © 1973 by John Wiley & Sons, Inc.
The Total Synthesis of Natural Products VOLUME 1
Edited by
JOHN ApSimon Department of Chemistry
Carleton tmiuersify, Ottawa
WILEY-INTERSCIENCE, a Division of John Wley & Sons, Inc. New York -.London .Sydney
Toronto
A NOTE TO THE READER This book has been electronically reproduced from digital information stored at John Wileydr Sons, Inc. We are pleased that the use of this new technology will enable us to keep works of enduring scholarly value in print as long as there is a reasonable demand for them. The content of this book is identical to previous printings.
Copyright @ 1973 , by John Wiley & Sons, Inc All rights reserved. Published simultaneously in Canada. Reproduction or translation o f any part of this work beyond that permitted by Sections 107 or 108 o f the 1976 United States Copyright Act without the permission of the copyright owner is unlawful. Requests for periiii~nionor further information should be addressed to thc Permissions Departnient. John Wiley & Sons. Inc.
Library of Congress Cataloging in Publication Data: ApSimon, John. The total synthesis of natural products.
Includes bibliographical references. 1. Chemistry, Organic-Synthesis. 1. Title. QD262.A68 547l.2 ISBN 0-471-03251-4
12-4075
Printed in the United States of America I0 9 8
Contributors to Volume 1 U. F. A. F. J.
Axen, Upjohn Company, Kalamazoo, Michigan M. Dean, University of Liverpool, England H. Jackson, University College, Cardiff, United Kingdom Johnson, Dow Chemical Company, Wayland, Massachusetts K. N. Jones, Queen’s University, Kingston, Ontario S. A. Narang, National Research Council of Canada, Ottawa J. E. Pike, Upjohn Company, Kalamazoo, Michigan W. P. Schneider, Upjohn Company, Kalamazoo, Michigan K. M. Smith, University of Liverpool, England W. A. Szarek, Queen’s University, Kingston, Ontario R. H. Wightman, Carleton University, Ottawa, Ontario
Preface Throughout the history of organic chemistry we find that the study of natural products frequently has provided the impetus for great advances. This is certainly true in total synthesis, where the desire to construct intricate and complex molecules has led to the demonstration of the organic chemist’s utmost ingenuity in the design of routes usingestablished reactions or in the production of new methods in order to achieve a specific transformation. These volumes draw together the reported total syntheses of various groups of natural products with commentary on the strategy involved with particular emphasis on any stereochemical control. No such compilation exists at present and we hope that these books will act as a definitive source book of the successful synthetic approaches reported to date. As such it will find use not only with the synthetic organic chemist but also perhaps with the organic chemist in general and the biochemist in his specific area of interest. One of the most promising areas for the future development of organic chemistry is synthesis. The lessons learned from the synthetic challenges presented by various natural products can serve as a basis for this everdeveloping area. It is hoped that this series will act as an inspiration for future challenges and outline the development of thought and concept in the area of organic synthesis. The project started modestly with an experiment in literature searching by a group of graduate students about six years ago. Each student prepared a summary in equation form of the reported total syntheses of various groups of natural products. It was my intention to collate this material and possibly publish it. During a sabbatical leave in Strasbourg in the year 1968-1969, I attempted to prepare a manuscript, but it soon became apparent that if 1 was to also enjoy other benefits of a sabbatical leave, the task would take many years. Several colleagues suggested that the value of such a collection vii
viii
Preface
would be enhanced by commentary. The only way to encompass the amount of data collected and the inclusion of some words was to persuade experts in the various areas to contribute. I am grateful to all the authors for their efforts in producing stimulating and definitive accounts of the total syntheses described to date in their particular areas. I would like to thank those students who enthusiastically accepted my suggestion several years ago and produced valuable collections of reported syntheses. They are Dr. Bill Court, Dr. Ferial Haque, Dr. Norman Hunter, Dr. Russ King, Dr. Jack Rosenfeld, Dr. Bill Wilson, Mr. Douglas Heggart, Mr.George Holland, and Mr. Don Todd. I also thank Professor Guy Ourisson for his hospitality during the seminal phases of this venture. JOHN APSIMON Ottawa, Canada Febrrtary 1972
Contents The Total Synthesis of Carbohydrates
1
J. K. N. JONESand W. A. SZAREK The Total Synthesis of Prostaglandins
u. AXBN,J.
E.
PIKE, AND
w. P. SCHNEIDER
The Total Synthesis of Pyrrole Pigments
81
143
A. H. JACKSON AND K. H. SMITH
The Total Synthesis of Nucleic Acids
s. A.
NARANG AND R. H.
279
WIGHTMAN
The Total Synthesis of Antibiotics
331
F. JOHNSON The Total Synthesis of Naturally Occurring Oxygen Ring Compounds 467 F. M. DEAN Subject Index
563
Reaction Index
597 ix
THE TOTAL SYNTHESIS OF NATURAL PRODUCTS
Total Synthesis OfNatural Products, Volume1 Edited by John Apsimon Copyright © 1973 by John Wiley & Sons, Inc.
The Total Synthesis of Carbohydrates J. K. N. JONES AND W.A. SZAREK
Department of Chemistry, Queen's University, Kingston, Ontario, Canada
1. Introduction
2. Base-Catalyzed Condensations with Carbon-Carbon Bond Formation. The
Formose Reaction 3. Syntheses from Acetylenic and Olefinic Precursors 4. Syntheses from Tartaric Acid and Other Naturally Occurring Acids 5. The Diels-Alder Reaction 6. Syntheses from Furan and Pyran Derivatives 7. Miscellaneous Syntheses A. Amino Sugar Derivatives B. Deoxyfluoro Sugar Derivatives C. Branched-Chain Sugars 8. Enzymic Syntheses 9. Synthesis of Cyclitols References
1 2 8 19 24 28 58 58
60
62 64
66 75
1. INTRODUCTION
The carbohydrates comprise one of the major classes of naturally occurring organic compounds. Although the structures of carbohydrates appear to be quite complex, the chemistry of these compounds usually involves only two kinds of functional group, ketone or aldehyde carbonyls and hydroxyl groups. The carbonyl groups normally are not free but are combined with 1
2
The Total Synthesis of Carbohydrates
the hydroxyl groups in hemiacetal or acetal linkages; the carbon of the “masked” carbonyl is known as the anomeric center. When a free sugar is dissolved in an appropriate solvent, a dynamic equilibrium is achieved involving both anomerization and ring isomerization. A wide variety of sugars has been found in nature and/or synthesized in the laboratory. These include not only the “classical” sugars but also derivatives such as amino, thio, halo, deoxy , branched-chain, and unsaturated sugars. Most synthetic sugars have been obtained by chemical transformations of naturally occurring sugars or their derivatives. In fact, the degree of achievement is such that the synthesis of a new mono-, di-, or trisaccharide can now be undertaken with a fair degree of confidence. The total synthesis of sugars from noncarbohydrate precursors has also been achieved by many routes. Some methods are long, involved, are stereospecific and result in the formation of one or two sugars only; others are relatively simple but produce complex mixtures of carbohydrates which may resist fractionation. Practically all naturally occurring sugars are optically active. Most synthetic routes which employ noncarbohydrate precursors produce racemic mixtures of sugars which may be difficult to separate into the D and L isomers. However, if enzymes are used to effect condensation of fragments or to remove one or more of the components, optically pure isomers may be isolated. In this chapter the total synthesis of sugars and the. related alditols and cyclitols, from noncarbohydrate substances by both specific and nonspecific methods, are discussed. Only compounds containing more than three carbon atoms are considered. 2. BASE-CATALYZED CONDENSATIONS WITH CARBON-CARBON BOND FORMATION. THE FORMOSE REACTION
The formose reaction has attracted the attention of biologists and chemists in recent years because it involves the self-condensation of formaldehyde to produce reducing sugars. This property is of interest in considering the problem of the origin of life on this planet, especially as formaldehyde has been detected in interstellar gases,‘ and also because of the feasibility of using carbon (as formaldehyde) as a possible source of sugars for the growth of microorganisms with the concomitant production of proteins and other complex organic compounds of importance to life and industry.2 The self-condensation of formaldehyde under the influence of base to yield a sugarlike syrup (methylenitan) was first observed by ButlerowS in 1861, when he treated trioxymethylene with calcium hydroxide solution. Calcium carbonate, magnesia, baryta, mineral clay, or even y-radiation
2. Base-Catalyzed Condensations
3
may also be used.4 Fischer5 showed that the sugarlike syrup called formose obtained from gaseous formaldehyde,“ or methose prepared by the action of magnesium hydroxide suspension on formaldehyde,’ contained monosaccharides, and that by reacting methose with phenylhydrazine acetate it was possible to obtain two hexose phenylosazones in low yield. Much higher yields were obtained from acrose (see below). Fischer named these derivatives a- and ,9-acrosazone (the corresponding sugars are a-acrose and 8-acrose) and showed that a-acrosazone was DL-arabino-hexose phenylosazone (DLglucose phenylosazone). In a remarkable series of experiments involving chemical and enzymic processes, Fischer was able to achieve the total synthesis from a-acrose of D- and L-glucose, D-gluconic acid, D- and Lmannose, D- and L-mannonic acids, D- and L-mannitol, and of D- and L-arabino-hexulose (D- and L-fructose), thus laying the basis for the synthesis of many other sugars. The transformations achieved are shown in Scheme 1. 8-Acrosazone was later identified as DL-xylo-hexose phenylosazone (DL-sorbose phenylosazone), which can be derived from DL-glucose, DL-idose, and ~~-xylo-hexulose (DL-sorbose).8 However, there has been a suggestion that ,9-acrosazone is really the phenylosazone of DL-dendroketose (see below). Fischer and TafeP observed that 2,3-dibromopropionaldehyde (acrolein dibromide) when treated with dilute alkali yielded products which reacted like sugars (hence the name “acrose”). DL-GlyceraldehydelO also gave products that possessed the properties of sugars when treated similarly. Fischer and TafelS explained the formation of acrose as an aldol-type condensation between DL-glyceraldehyde and lY3-dihydroxypropanone (dihydroxyacetone), the latter compound being formed by a base-catalyzed isomerization from DL-glyceraldehyde (see Scheme 2). H. 0.L. Fischer and E. B a e P showed that D-glyceraldehyde and 1,3dihydroxypropanone react in basic solution to yield D-arabino-hexulose and D-xylo-hexulose as major products. This reaction is a general reaction and novel sugars can be produced if D-glyceraldehyde is replaced by L-glyceraldehyde or by other aldehydes (see below). It is interesting that the biological origin of D-arabino-hexulose follows a similar route, but sugar phosphates and enzymes (aldolases) are involved.12 In all cases the threo configuration is favored at the newly formed asymmetric centres but condensations involving enzyme-catalyzed reactions13 usually yield the D-threo configuration only. The conversion of formaldehyde to formose involves a complex series of reactions which have been rationalized by Breslow,14 who suggested that two processes are involved in the formation of glycolaldehyde from formaldehyde. The first is a slow condensation of two molecules of formaldehyde to form glycolaldehyde, which then reacts rapidly with a further molecule of formaldehyde to produce glyceraldehyde. Part of this is then converted to
P
4
R
R
I HOCH I
COOH
D-arubino-Hexose phenylosazone
conc. HCI
R
D-Mannose
I
HC-
Scheme 1
I
1
zn
yeast
I I
L-Fructose
(only the D isomers of DL mixtures are shown)
CHzOH
I
HCOH
I
HCOH
R = HOCH
DL-Mannitol
I
R
CH,OH
lNa-Hn
I a HOCH
D-Fructose
r-Mannose
yeast
DL-Mannose
R
R
I
Dr-Fructose
AcCH
1
C=O
CH,OH
- Zn
(DL-urubino-Hexosulose, DL-Glucosone)
D-arabino-Hexosulose
-
I I
HC=O HOCH
\Na-Hn
~-Gluconicacid
R
I I
COOH
auinoline A
R
a-Acrosone
Z H O C H
D-Mannonic acid Dr-Mannonic acid
Na-Hg
PhNHNHz
i
D-Glucose
I
resolved via strychnine sal~s
HCOH -HCOH
I I
HC=O
L-Mannonic acid
R
H G O
I conc.HCI I C=NNHPh + C=O I I
HC=NNHPh
a-Acrosazone (DL-urubino-Hexose phenylosazone
Formaldehyde, PhNHNH, glyceraldehyde or + a-Acrose * acrolein dibromide
2. Base-Catalyzed Condensations
HC=O
I CHBr I
HC=O
+
CH,Br CHZOH
I
C=O I CHzOH
HCOH
I II CHOH e COH I I
one
CHzOH
+
CHZOH
I e C=O
I
CH,OH
HC=O
5
CHzOH
CH,OH
I
I
OH'
CHOH --+
C=O
CH,OH
CHOH
I
HZ0
I
I I CHOH I CHOH
CH,OH Scheme 2
1,3-dihydroxypropanone,* which then rapidly reacts with formaldehyde to yield tetrulose and then tetrose, which then breaks down to two molecules of glycolaldehyde. The reaction is thus autocatalytic and is formulated as shown in Scheme 3. The rate of formose forfiation is dependent upon the metal cation of the base used. It is more rapid with those bases that form chelate compounds with enediols, which are intermediates in the foregoing reaction: thallium hydroxide > calcium hydroxide > sodium hydroxide. It follows that the composition of formose will depend upon the base used, the concentration of the reactants, and the temperature and time of reaction. Short periods of reaction favor the formation of lower molecular weight ketose sugars, longer periods of reaction yield more aldose sugars, while high concentration of alkali and long periods of heating yield saccharinic acids16 and other products resulting from the decomposition of sugars by
* 1,3-Dihydroxypropanone has been prepared by Marei and RaphaelI5 from nitromethane and formaldehyde: CH,NO,
+xH,o
CH,OH OH'
---+H O C H , - - L O , I
CHZOH
CH,OH PhCHO A
H@
OCH
I
2-
HC-OCH,
Ph
I
C-NO,
I
1..
CHZOH HOCH 2-
C-CHZOH
I1
0
1) NaIO, t-2)H@
I
OCHZ-C-NH,
I
HC-OCH, Ph
I
6
The Total Synthesis of Carbohydrates
CH2O CHZO
+ HOCH2-CHO
+ HOCH,-CO-CH2OH
HOCHZ-CHOH-CHOH-CHO
3 HOCHZ-CHOH-CHO
Z
HOCH2-CO-CHzOH Z HOCH2-CHOH-CO-CH2OH Z HOCH2-CHOH-CHOH-CHO ;+ 2 HOCHa-CHO Scheme 3
alkali. With the advent of paper chromatography and gas-liquid chromatography, it has been possible to detect all eight aldohexose sugars, all four hexuloses, the four pentoses, two pentuloses, all possible tetroses, dendroketose, and three h e p t u l o s e ~ . ~Recently, ~-~~ sugars prepared by base-catalyzed condensation of formaldehyde were analyzed by combined gas-liquid chromatography and mass spectrometry; both branched and straight-chain products weredetected.lgaCannizzaro reaction of formaldehyde proceeds in alkaline medium in conjunction with the formose reaction to produce aldoses and ketoses, and it has been shown1Bbthat the extent of the two reactions is a function of the catalyst used. In a study with calcium hydroxide as catalyst, it was found that the ratio of branched-chain sugar derivatives, such as (hydroxymethy1)glyceraldehydeand apiose (see below), and straight-chain products could be controlled by manipulation of the reaction conditions. The branched products are very readily reduced by a crossed-Cannizzaro reaction with formaldehyde and large quantities of species such as (hydroxymethy1)glycerol are produced. Formose solutions are decomposed by microorganisms if allowed to stand in an open vessel in the laboratory.20 Glycolaldehyde itself polymerizes under the influence of base to yield tetroses, hexoses, and other sugars.21 Methoxyacetaldehyde polymerizes in aqueous potassium cyanide solution forming 2,4-dimethoxyaldotetroses :22 KCN CH3OCH2-CHO CHSOCHZ-CHO or
+
KaC05
CHSOCHZ-CHOH-CHOCHS-CHO The polymerization of formaldehyde to yield sugars is, therefore, a very complicated process. For example, the formation of pentoses from formaldehyde may proceed via several routes. Glycolaldehyde and 1,3-dihydroxypropanone may react to form pentuloses, which subsequently are isomerized to pentoses, or formaldehyde and a tetrulose may combine to yield pent-3uloses, which then isomerize to pentuloses and pentoses, or glyceraldehyde and glycolaldehyde may combine to form pentoses. To test these hypotheses, Hough and Jonesz3 treated mixtures of glycolaldehyde and 1,3-dihydroxypropane and of glyceraldehyde and glycolaldehyde with lime water and found that pentoses were produced along with several other sugars. They were able to isolate arabinose, ribose, and xylose, as phenylhydrazones, from the complex mixture of sugars that results from the two reactions previously described.
2. Base-Catalyzed Condensations
7
Very recently, it was that it is possible, by making use of the hexokinase reaction, to extract some specific sugars from the complex synthetic formose sugars. The enzyme hexokinase is known to transfer the terminal phosphate of ATP to D-ghIcose: D-glucose
+ ATP - D-glucose 6-phosphate + ADP hexokinase
However, the enzyme is not totally specific for glucose, other hexoses such as fructose and mannose being also susceptible to phosphorylation. The basis of the method of extraction involves phosphorylation of some hexoses by this means, which are then retained on a column of anion-exchange resin (together with unreacted ATP and formed ADP), while other unreacted, neutral components of the formose mixture pass through the column. The sugar phosphates are then eluted by a salt solution of appropriate concentration, and the unsubstituted hexoses are obtained by a phosphatase reaction. The branched-chai,n sugar DL-dendroketose mentioned earlier was first isolated by Utkin,24 who prepared it by adding sodium hydroxide to a solution of 1,3-dihydroxypropanone in water. It is formed so easily and in such high yield that it seems remarkable that it has not appeared in any natural product. Moreover, it is metabolized completely by baker's yeasLZ6 Like the branched-chain sugar apiose,26hemiacetal formation results in the formation of a new optically active center with the possible formation of eight isomers from the D and L forms of dendroketose. Utkin was able to isolate D-dendroketose (4-C-hydroxymethyl-~-glycero-pentulose) when he observed that a microorganism which accidentally contaminated a solution of DLdendroketose, metabolized the L-isomer only. He was able to prove the absolute configuration of the nonmetabolized material by relating it to ~-apiose,Z'a sugar of known absolute configuration, by the series of reactions indicated in Scheme 4. It may be significant that D-dendroketose, which remained after fermentation of the DL mixture, possesses a potential L-threo disposition of hydroxyl groups at C-3 and C-4, while L-dendroketose which possesses a potential D-rltreo configuration at C-3 and C-4 is metabolized: CHzOH
I c=o I I
HOCH ~~
HOCHZ-C-OH
I
CH,OH L-Dendroketose (metabolized)
CHaOH
I c=o
I I HO-C-CH20H I HCOH
CH,OH
D-Dendroketose (not metabolized)
8
The Total Synthesis of Carbohydrates
CH20H
CH,OH
I c=o
7 I
I I
CHOH
HCOH
I
/ \
t
I I
HCOH C-OH
/ \
CH20H CH20H
CHzOH CH20H
DL-Dendroketose
D-Dendroketose
CH20H
CO,H
I
fermentation
C-OH
/3 HOCHZ-C \
I c-=-o
c=o
H-CHOH
HOO
CH,OH
I
HC=O 8%
t
C-OH
/ \
I I
HCOH C-OH
/ \
CHZOH CHzOH CH2OH CHZOH D-Apionic acid
Me,
o-Apiose
0,
c
P
Scheme 4
3. SYNTHESES FROM ACETYLENIC AND OLEFINIC PRECURSORS
The directed synthesis of carbohydrates from noncarbohydrate precursors in most cases involves the preparation of compounds of acetylene. These acetylenic intermediates may be converted into cis- or trans-ethylenic derivatives dependent upon the mode of reduction of the acetylene. A further advantage of this approach is that the ethylene may then be hydroxylated in a cis or trans fashion, as decided by the mode of oxidation. In some cases steric effects may be used to force the predominant formation of one of the DL forms. This procedure is particularly effective when the hydroxylation of a ring compound is involved.
3. Synthesis from Acetylenic and Olefinic Precursors
9
Several workers, chief among whom are Lespieau, Iwai, and Raphael, have synthesized carbohydrate derivatives from acetylenic and olefinic precursors. Stereochemical problems of hydroxylation were minimized either by cis-hydroxylation of double bonds of known stereochemistry using potassium permanganate or osmium tetroxide, or by epoxidation of double bonds of known stereochemistry followed by opening of the epoxide ring, with resulting trans-hydroxylation of the double bond. Griner2a appears to be one of the first to attempt the synthesis of sugar alcohols. He observed that when acrolein was hydrogenated by means of a zinc-copper couple and acetic acid, dimerization occurred and divinylglycol (CH,=CH-CHOH-CHOH-CH=CH,) resulted. This may exist in niem or DL modifications. Griner obtained the aid of LeBel to isolate a mold which would preferentially metabolize one of the isomers. I n this, LeBel was successful. Griner had expected to obtain an optically active material but obtained a product devoid of activity and concluded that the nieso form only was present. L e s p i e a ~later ~ ~ showed this conclusion to be erroneous, Griner attempted to oxidize the divinylglycol, with permanganate solution, to a hexitol, but was unsuccessful. Later, in a brief note,3O Griner stated that addition of two molecules of hypochlorous acid to divinylglycol gave a divinylglycol dichlorohydrin from which, after treatment with base, he was able to isolate m-mannitol. Lespieau3I repeated the attempted hydroxylation of divinylglycol but used osmium tetroxide-silver chlorate as the hydroxylating agent, and obtained allitol and m-mannitol (see Scheme 5).
2CH2=CH-CHO CH20H
I I HCoH AgCIOS - I I HCOH OSo4 I HCOH I HCOH
CHzOH Allitol
CH2
pz?
CH,
CH,OH
II I HOCH CH I I HCOH HOCH AgC,OQ HOCH + HCOH I I HCOH HCOH I I I CH HCOH CH II I II
II CH I
CH2
ttieso
CH2
D-t/itZO
+
( L-//ireo)
1
I ) HOCl
2) @OH
DL-Mannitol Scheme 5
CH20H
\
CH20H
I I HCOH I HOCH I HOCH I HCOH
CH20H DL-Mannitol /
10
The Total Synthesis of Carbohydrates
Hence, assuming cis addition of the new hydroxyl groups, allitol arises from the meso compound and ~ ~ - m a n n i t ofrom l Dbdivinylglycol. In a second method of synthesis,32involving the Grignard reagent derived from acetylene and chloroacetaldehyde, divinylacetylene dichlorohydrin
(CH,CI-CHOH-C-C-CHOH-CH~Cl), was prepared, converted to the hexynetetrol, and reduced to the corresponding ethylene derivative. Hydroxylation of the product by means of osmium tetroxide-silver chlorate gave galactitol and allitol. The ethylene derivative, therefore, had the meso configuration (see Scheme 6). Lespiead3 also synthesized ribitol and DL-arabinitol using acrolein dichloride and acetylene as starting materials as shown in Scheme 7. Raphael34 improved on these syntheses by using epichlorohydrin and acetylene as starting materials, and performic acid as the oxidizing agent (see Scheme 8). CH,CI-CHO
+ BrMgCrCMgBr + CH,CI-CHO
1
I
CH2CI-CHOH-C-C-CHOH-CH,CI KOH ( v h diepoxy derivative)
CH,OH-CHOH-C-C-CHOH-CH,OH
1"
Pi-on-starch (Bourguel's C ~ I ~ ~ Y S I )
H H H H CH2OH-C-C=C-C-CH,OH AaCIOS 10SO.
CH,OH
CH,OH
I
1
HCOH
HCOH
I
I
HCOH
I I HCOH I
+
HCOH
HOCH
I
HOCH
I I
HCOH
CH20H Allitol
CHzOH Calactitol
&heme 6
\
-
m-Arabinitol pentaacetate
/
CH,OAc
I
AcOCH
HCOAc
I I AcOCH I
CH,OAc
+
I I
+
CH,OAc
C'Z
I
I
Ribitol pentaacetate Scheme 7
CHZOAC
')%?O,
H2
CH,OAc
CH,OAc
I
I
I
CHOAc CHOAc
I
t--CHOAc
CHOAc
111 I
C
CH
I! I
CH
CH2
' I
I
CHZCI
I
I
I
CHOH
CH,CI
t
CHOH
I
KOH (via epoxidc)
111 C
CH
&HCI
I
+ CHOH
I
C
CH~OAC
I
I
HCOAc
HCOAc
I
+ HCOAc
HCOAc
I I
HC=CMMnBr
CH,OAc
kH,CI
I
CHCl
HG=O
HCOAc
AcOCH
8Hz
I1
CH
I
HG==
Ill
CH
12
The Total Synthesis of Carbohydrates
CH,CI
I
I\
HC
- -
CH
C
C
Ill
Ill
I
HCECNa
0
CH
HCO,H
CH
II
CH
I
CH~OH
CH,Br
I HCOH I HCoAc I HCOAc I
I
CHOH
I
CHOH I CH20H
CH2Br
I
HOCH
+
CH,OAc
I
I
HCoAc
I
HCOAc
-1
N-bromosuccinimide
I
CH,OAc
CHOAc CHOAc
I
CH,OAc
D-ribo ( +L-ribo) D-arabirio (+L-arubilro) IA'OAC
Ribitol penlaacetate
lA80AC
DL-Arabinitol penlaacetate
Scheme 8
Iwai and his associates in Japan have achieved several total syntheses of pentose sugars using acetylenic compounds. Iwai and I w a ~ h i g condensed e~~ the Grignard reagent derived from 3-(tetrahydropyranyL2'-0xy)-propyne with 2,2-diethoxyacetaldehyde to yield 1,l-diethoxy-5-(tetrahydropyranyl2'-oxy)pent-3-yn-2-01, which, on reduction with lithium aluminum hydride, yields the wans-olefin. Catalytic hydrogenation, on the other hand, yields the cis-olefin. Acetylation of these products, followed by cis hydroxylation of the double bonds, affords products which, after hydrolysis of the acetal residues, yield the four DL-pentose sugars (see Scheme 9). Iwai and Tomita have achieved a stereospecific synthesis of DL-arabinoseS6 and a synthesis of a mixture3' of DL-arabinose and DL-ribose as shown in Scheme 10. DL-Ribose has been ~ynthesized~~ stereospecifically by oxidative hydroxylation of 2-ethoxy-5-(tetrahydropyranyl-2'-oxy)methyl-2,5-dihydrofuran (see Scheme 1 I), which was obtained by hydrogenation of DL-1,1-diethoxy-5(tetrahydropyranyl-2'-oxy)pent-2-yn-4-ol.This acetylenic compound was prepared by the Grignard reaction of (tetrahydropyranyl-2'-oxy)acetaldehyde with propargyl diethyl acetal magnesium bromide. One method for the
3. Syntheses from Acetylenic and Olefinic Precursors
+ OHC-CH(OEt),
ROH,C-C=CMgBr
.1
ROHZC-CrC-CHOH-CH(0E
.
r ROHZC
\
t),
LIAIH,
H
catalyst (Pd-CaCO,)
CHOH-CH(0Et)z
/
c=c
\ c=c/ / \
ROCHg
HC=O
I HCOH I HCOH I HCOH I
13
HC=O
HC=O
I HOCH
+
I HCOH
I
I
HOCH
HCOH
t.I
I
H OH
HCOH
I
CHzOH
CHzOH
DL-Ribose
DL-Arabinose
CHzOH
DL-XylOSe
CHOH-CH(OEt),
H
HC=O
I
HOCH +HOCH
I
HCOH
I
CH,OH DL-Lyxose
Scheme 9
preparation of (tetrahydropyranyl-2’-oxy)acetaldehyde involved ozonolysis of the tetrahydropyranyl ether of ally1 alcohol. This synthesis of DL-ribose is the first example, in this chapter, of a total synthesis of a sugar, which involved a furan derivative; other examples are discussed in Section 6. Total syntheses of deoxypentose sugars have also been reported. Hough30 described a preparation of the biologically important 2-deoxy-~-erythropentose (2-deoxy-~-ribose)which involves the reaction of 2,3-O-isopropylidene-D-glyceraldehyde with allylmagnesium bromide. Hydroxylation ofthe resultant 5,6-O-isopropylidene-I -hexene-~-erythro-4,5,6-triol gave a mixture of products. Periodate oxidation of the hexitol derivatives, followed by hydrolysis, afforded almost exclusively 2-deoxy-~-erythro-pentose(Scheme 12). Another preparation of this sugar has also been achievedQousing 2,3-0isopropylidene-D-glyceraldehyde as a starting material, by condensation
14
The Total Synthesis of Carbohydrates
HOCHZ-C=C-CHzOH
-1
POCI,
CICHZ-CGC-CH~CI
1 1
EtONa
EtO-CH=CH-C=CH 1) ElM8Br in THF
2) HCHO
EtO-CH=CH-C-C-CH~OH
I
EtO-CH=CH-CEC-C
HZOAc
EtO-CH=CH
JLiAlH4
i
H,, Pd-CaCO,
EtO-CH=CH
I
\ H
H EtO-CH=CH /CH20Ac
c=c
/
\
\
/
H
c=c
\
CH20H
\ / c=c / \
H
H
1. KMnO,
/
H
CH,OAc
1i:T
2. H @
HC=O
HC=O
I HOCH I HCOH I HCOH I
I
HCOH
DL-Arabinose
I + HCOH I I
HCOH
CHaOH
CHzOH
DL-Ara binose
DL-Ribose
Scheme 10
with acetaldehyde in the presence of anhydrous potassium carbonate; 2-deoxy-~-xylosewas also obtained. Fraser and Raphaelq1have synthesized 2-deoxy-~~-erythro-pentose from but-2-yne-1,Cdiol (Scheme 13). This compound was converted into 1benzoyloxy-4-bromobut-2-yne (1) by monobenzoylation and treatment of the resultant half-ester with phosphorus tribromide. Condensation of
3. Syntheses from Acetylenic and Oleflnic Precursors
HOCHz-CH=CHz
15
+ ROCH2-CHO
----+ ROCHz-CH=CH2
B~ME-C=C-CH(OC~H~)~J 4
Ha. Pd-CaCO,
ROCHz-CH-C~C-CH(OC2H,),
I
ROCHz=0c2,,,
OH
1
KMnOd
b
ROCH,
OC2145 H@ --f
DL-Ribose
OH OH
Scheme 11
l-benzoyloxy-4-bromobut-2-ynewith ethyl sodiomalonate gave ethyl 5benzoyloxypent-3-yne-1,l-dicarboyxlate(2), which was converted into the dihydrazide 3. Compound 3 was then subjected to a double Curtius rearrangement; reaction with nitrous acid, followed by treatment of the resultant diazide with ethanol, afforded the acetylenic diurethane 4. Catalytic hemihydrogenation of 4 gave the cis-ethylenic diurethane 5. cis-Hydroxylation of 5, followed by acid-catalyzed hydrolysis of the resultant erythrotriol, gave finally a small yield of 2-deoxy-~~-erythro-pentose. Weygand and Leube4, have also prepared 2-deoxy-~~-erytfrro-pentose (and 2-deoxy-~~-threo-pentoseor 2-deoxy-~~-xylose) from an acetylenic precursor, 1-methoxy-1-buten-3-yne(6). Treatment of 6 with formaldehyde CH2
II
CH
I
CHzMgBr
+
HC=O
I
HCO HzCO
CHZ
II CH I CHZ I
+ HCOH
I
HCO
CH,OH
HC=O
I
I I HCOH I HCOH
- CHOH
I
CH,
H A
I-BuOH
I
1) NaIO,
I
2)H'
HCOH
080,
HCO
\
H2C0 Scheme 12
CH2
I
CHzOH 2-DeoxyD-erytftrO-
pentose
16
The Total Syn!hesis of Cnrbohydrates
HOCHz-C=C-CH20H
--f
PhCOCHz-C=C-CHzBr
I1
0 1
J PhCOCHz-CrC-CH,-CH(COzEt)z II
2
HOCHz-CrC-CHz-CH(CNHNHJ2
1I
0
3
HOCHz-C=C-CHz-CH(
NH-COZEt),
4
HOCHZ--C=C-CH,-CH(NH-COzEt)z H H
I 5
1) KMnO, 2) He
2-De~xy-o~-erytltro-pentose
Scheme 13
in methanol at 45-55" gives l-methoxy-l-penten-3-yn-5-01, but at 65-85" the dimethyl acetal 7 was produced. Hemihydrogenation of 7 over a Lindlar catalyst gave the corresponding ethylene 8. Hydroxylation of 8 with osmium tetroxide and hydrogen peroxide in t-butanol, followed by acid-catalyzed whereas the use of peroxyhydrolysis, gave 2-deoxy-~~-erythro-pentose, (see Scheme 14). benzoic acid afforded 2-deoxy-~~-iltreo-pentose A more recent synthesis of 2 - d e o x y - ~ ~and - L-erythro-pentose has been reported by Nakarninami et The first step (see Scheme 15) was a Reformatsky reaction of ethyl bromoacetate with acrolein to give the 8-hydroxy ester 9. Compound 9 was hydrolyzed by aqueous potassium hydroxide to give the DL-acid 10, which was treated in an aqueous solution
3. Syntheses from Acetylenic and OleBnic Precursors
HCEK!-CH=CHOCH,
--t
6
17
HOCH~-CSC-CH~-CH(OCHJ, 7
J
HOCHz-C=C-CH2-CH(OCH3)2 H H 8
I
1) H,O,, OsO, or KMnO, 2) H@
2-Deoxy-~L-erythro-pentose
0 1) PhCOOH
”
2) H@
2-Deoxy-~~-threo-pentose
Scheme 14
with N-bromosuccinimide to afford the DL-bromolactone 11. Successive (12). Treatment basic and acidic hydrolysis gave “2-deoxy-~~-ribonolactone” of 12 with disiamylborane [bis(3-methyl-2-butyI)borane], and hydrolysis of the resultant tris(disiamy1borinate) ester, yielded 2-deoxy-~~-eryt/troC02Et
C0,Et
I CH2Br
+
HC=O
I CH
-%
I CH2 I
CHOH
C02H
I I
CH2 KOH __t
CHOH
5 Ha0
HOCH O = I V HC0
I
CH2Br DL-rhreo
9
10
c=o
-I
I
CHZ I ) disiamylboranc H,O
2 Deoxy-DL-erythro-pentose <2)
I
HOCH 0-CH
Scheme 15
I
18
The Total Synthesis of Carbohydrates
pentose. The preparation of 2-deoxy-~-erythro-pentoseinvolved treatment of the racemic hydroxy acid 10 with a half equivalent of quinine and decomposition of the salt to yield (-)-lo, which was then subjected to the same reactions as in the case of the racemic compounds. 2,3-Dideoxy-~~-pentose has been synthesized by Price and B a l ~ l e yby ~~ the Claisen rearrangement of ally1 vinyl ether to 4-pentenal, conversion to the methyl acetal, and permanganate oxidation (see Scheme 16). Total syntheses of tetroses and tetritols from olefinic precursors have also been achieved, using reactions which have already been described in this section. Thus RaphaeP5 obtained erythritol tetraacetate by treatment of rrans-Zbutene-l,6diol diacetate (13) with peroxyacetic acid, followed by complete acetylation, whereas similar treatment of the cis compound 14 produced DL-threitol tetraacetate: AcOCH,
\
H
/
H
AcOCH,
\
c=c
/
\
H
CH,OAc
13
c-c
/
/cH20Ac
\
H
14
As usual, cis-hydroxylation of the cis-diacetate yielded erythritol tetraacetate, after complete acetylation, and the trans-diacetate gave DL-threitol tetraacetate, on treatment with osmium tetroxide and hydrogen peroxide in CHz=CHOCH,CH=CH,
265'
I
+CH,=CHCH,CH,CHO CH,OH CsCl,
KMnO
HOCH,CHCH2CH2CH(OCHJ, LCH,=CHCH,CH,CH(OCH,)z I OH
\
HC=O
I I
CH, CH2
I I
CHOH CH,OH Scheme 16
4. Syntheses from Tartaric Acid
19
t-butanol and complete acetylation of the product. trans-Addition of hypobromous acid to the cis and trans compounds (14 and 13) was smoothly effected to give the threo- (15) and erythro-2-bromobutane-1,3,4-triol 1 ,Cdiacetates (see Scheme 17). Chromium trioxide oxidation of either the eryrhro- or the threo-bromohydrin gave the same ketone, 2-bromo-l,4diacetoxybutan-3-one (la), which, on treatment with silver acetate in acetic acid, yielded DL-glycero-tetrulose triacetate. Hydrolysis with baryta then gave DL-glycero-tetrulose. Other examples of the preparation of tetritol derivatives from olefinic precursors are known.45a Kiss and Sir~kmin*~’’ synthesized erythro-2-amino-l,3,4-trihydroxybutanestereospecifically from trans-1,4-di bromo-2-butene.
H
14
HOBr
H O AcOCH~-C-C-CH~OAC Br H DL-threo
15
H AcOCH2-C-C-CH,0AC
.=h
Br II
J
16
0
1) AnOAc-AcOH
2) @OH H HOCHz-C-C-CH2OH
OH II
0
oL-(ilycero-tetruloa
Scheme 17
Waltonq6has prepared D-threose and D-erythrose by a method similar to that employed by Hough3# for the synthesis of 2-deoxy-~-erythro-pentose. Thus addition of vinylmagnesium chloride to 2,3-O-isopropylidene-a/dehydoD-glyceraldehyde gave a mixture of epimeric pentene derivatives, which were separated by gas-liquid chromatography. Ozonolysis followed by acid-catalyzed hydrolysis of each epimer afforded D-threose and D-erythrose, each in approximately 40% yield. A similar study has been made by Horton et a1.4’ In that work, however, the first step was ethynylation of 2.3-0isopropylidene-aldehyde-D-glyceraldehyde to give a 44 :56 mixture of 4,5-O-isopropylidene-l -pentyne-D-erythro (and ~-tltreo)-3,4,5-triol. 4. SYNTHESES FROM TARTARIC ACID AND OTHER NATURALLY OCCURRING ACIDS
The potential of the tartaric acids as possible precursors in the synthesis of tetroses and related compounds was recognized by Emil Fischer as early
The Total Synthesis of Carbohydrates
20
HC=O
I I
HCOAc AcOCH
I
O=
CO,CH,
21
0
‘3 I1
COOH
I
HCOAc HCOH HOCH I ---+AcO{, Ac,O
COCl
I)CH,OH 2) SOCl,
C
LOOH
I HCOAc I
AcoSH
II
I
HCOAC I
BF,-ctherate -t ElOH
I
HCOAc I
ACO~H
I
I
CO,CH,
COZCH3 22
25
EtS
I c===o I
AcOCH
I
C0,CH3 24
-
-CI
SEt
COZCH, 19
R~~~~
I
I
c==o
HCOAc
(LI
AcO H CO,CH, 23
1 SEt ’
CO,CH, I CH, hv
I
HCOAc I
ACOCH
CO,CH, I 26
EtSCI
\ /
‘CL
CH3
HCOAc
t AcOCH
I
2) Ac,O
CHN, I C==O
A ~ O ~ H
I
I ) LiAIH[OC(CH,),1,.O0
17
CH,OEt I c- 0
I I
HCOAc
COZCH, 18
0
L-Tartaric Acid
CHZOAC
NaSEl t
I I AcOCH I
HCOAc
Scheme 18
as 1889;48 however, he was unsuccessful in attempts to reduce tartaric acid. In 1941 Lucas and B a ~ m g a r t e nreported ~~ a solution to this problem, and achieved a synthesis of L-threitol. More recently, Bestmann and SchmiechenS0 employed L-tartaric acid for the synthesis of a variety of tetrose and pentose derivatives (see Scheme 18). A key intermediate in that workso was the acid chloride of monomethyl di-0-acetyl-L-tartrate (18). Compound 18 was also ~ ~preparation an intermediate in the work of Lucas and B a ~ m g a r t e n .Its involved heating L-tartaric acid with acetic anhydride to give di-0-acetyltartaric anhydride (17), which reacts vigorously with methanol to give
4. Syntheses from Tartaric Acid
21
monomethyl di-0-acetyltartrate; the latter compound was then converted into 18 by treatment with thionyl chloride. The acid chloride group in 18 was reduced to a hydroxymethyl group by Bestmann and Schmiechen, on treatment with lithium tri-t-butoxyaluminum hydride at 0"; the reduced product was isolated as methyl tri-0-acetyl-L-threonate (19), which was converted into L-threono-1 ,44actone (20). When the reduction of 18 with lithium tri-t-butoxyaluminum hydride was performed at -75", methyl 2,3-di-O-acetyl-~-threuronate (21) was produced; Lucus and Ba~mgarten*~ had obtained this compound by a Rosenmund reduction of 18.* The acid chloride 18 could be transformed with diazomethane into the Compound 22 was converted by Bestmann and Schmiediazoketone 22.61*6a chen50 into the diethyl dithioacetal 23 by a reaction with ethylsulfenyl chloride, followed by treatment of the intermediate I-chloro-1-ethylthio derivative with sodium thioethoxide. Desulfurization of 23 with Raney (24). nickel gave methyl 2,3-di-O-acetyl-5-deoxy-~-threo-4-pentulosonate Treatment of the diazoketone 22 with boron trifluoride-etherate in ethanol afforded methyl 2,3-di-O-acetyl-5-O-ethyl-~-t/~reo-4-pentulosonate (25). Compound 25 had been obtained earlier by UItCe and S O O ~ by S , treatment ~~ of 22 with cupric oxide in ethanol, instead of the expected Wolff rearrangement product. A Wolff rearrangement of the diazoketone 22 was achieved by Bestmann and Schmiechen by irradiation with ultraviolet light of a methanol solution of 22; the product was di-O-acetyl-2-deoxy-~-threopentaric acid dimethyl ester (26). The diazoketone 22 has also been utilized in a synthesis of the branchedchain sugar (see also Section 7C) L-apiose by Weygand and Schmiechensl (Scheme 19). Treatment of 22 with acetic acid in the presence of copper powder gave methyl 2,3,5-tri-O-acetyl-4-pentulosonate(27), which was converted into methyl 2,3,5-tri-O-acetyl-4,4'-anhydro-4-C-hydroxymethylL-tkreo-pentonate (28) with diazomethane. The opening of the epoxide ring, after saponification of 28 to give 29, was achieved with a strongly acidic ion-exchange resin; the resultant product (30) was finally converted into L-apiose by a Ruff degradation procedure (oxidative decarboxylation of the calcium salt of 30 with hydrogen peroxide in the presence of ferric acetate). Some deoxy sugar derivatives have been obtained by Lukes et al.53from
* Reduction of methyl 2,3-di-O-acetyl-~-threuronate(21) with sodium amalgam gave L-threonic acid, which was characterized as the brucine salt :4B CH,OH 21
Ns-Hs __f
I 1
HCOH
HOCH
I
COOH
22
The Total Synthesis of Carbohydrates
C02CH3
C02CH3
C02CH3
I
1 HCOAc I
I I
HCOAc
HCOAc
AcOCH
I
I
I
--+ AcOCH
AcOCH
---t
c=o
C=O
C-
CHN,
CH20Ac
CH~OAC
I
I
CH2
I
27
22
28
CO,H
CO2H
HCOH
HCOH
I
HC==O
I HOCH I
I
/Yo"
t HOCH
HOCH
t
CHZOH CH2OH
I I
I/O\
/t?
C-
I
CH2OH CHSOH
L-Apiose
CH,
CHzOH
30
29
Scheme 19
L-parasorbic acid (31), which was isolated from Sorbus aucuparia berries. Hydroxylation of 31 with osmium tetroxide and sodium chlorate gave 4,6-dideoxy-~-ribo-hexonic acid 1,Slactone (32). The calcium salt of the acid was then subjected to a Ruff degradation to afford 3,5-dideoxy-~erythro-pentose (33) (Scheme 20). A study, similar to the foregoing, is the stereospecific trans-hydroxylation of angelactinic acid (34)."*b6Jary and Kefurts6found that hydroxylation of HC=O
I I
HOCH
1
----+
-
1
CH2
I I
I
CH3
CH3 31
32
Scheme 20
33
4. Syntheses from Tartaric Acid
23
34 with peroxyacetic acid gave the lactones of 5-deoxy-~~-arabinonic acid and 5-deoxy-~~-ribonic acid (35 and 36) in the ratio of 2.8: 1. These compounds were converted into 1-deoxy-DL-lyxitol (37) and 1-deoxy-DL-ribitol (38) by treatment of the lactones with lithium aluminum hydride in tetrahydrofuran (see Scheme 21 ; only one isomer of DL mixtures is shown). Very recently, Koga et aLS6described a new synthesis of D-ribose from L-glutamic acid (39) without the necessity of resolution at intermediate
COOH
I CHOH I
HC
II I
CH CH3 34
----+
o=co=cq I HOCH
I
HYH HC
I
0
1+
CH3
I I
HCOH
HYH HC
I
I
CH3 36
35
4
CH3
I HOCH I HOCH I HCOH I
0
+
CH,OH 37
CH3
I
HOCH
I I HOCH I HOCH
CH,OH 38
Scheme 21
stages; the asymmetric center of 39 became C-4 in D-ribose (see Scheme 22). The amino acid was deaminated to give, after esterification, the lactone ester 40; this deamination was considered to proceed with full retention of configuration, because of the participation of the neighboring carboxyl group. Reduction of 40 with sodium borohydride in ethanol afforded the lactone alcohol 41, which was converted into the benzyl ether 42. Treatment of 42 with sodium and ethyl formate in ether gave the sodium salt 43, which, on being heated in acidic aqueous dioxane, afforded 5-0-benzyl-2,3-dideoxyD-pentofuranose (44), as a result of hydrolysis of the lactone ring, decarboxylation, and subsequent ring closure. Compound 44 was converted into a mixture of glycosides (see Section 6) which, on treatment with bromine and calcium carbonate, gave the monobromo derivative 45 as a mixture of
The Total Synthesis of Carbohydrates
24
diastereomers. Base-catalyzed dehydrobromination of 45 afforded the unsaturated derivative 46. Surprisingly, hydroxylation of 46 with potassium permanganate or with osmium tetroxide gave a mixture of methyl 5 - 0 benzyl-/?-D-ribofuranoside (47) and methyl 5-O-benzyl-a-~-lyxofuranoside (48). Compounds 47 and 48 could be separated as their acetonides or diacetates; alternatively, D-ribose could be isolated as its “anilide” by hydrogenation of the hydroxylation product to remove the benzyl group, followed by acid-catalyzed hydrolysis, and then treatment with aniline. COOH -
COOEt --
-H , N - C-~ H --
----
+
0-C-H CH,
CH,
I
CHZ-COOH
O=
39
I’hCH~OC‘H2
1;
EtooCb b== ROCH,
G
-CH,
40
--+
0
*--
Bs
-
I
4lR=H 42 R = CHaPh
I’hCHzOCH, bohczck
Q0CH3
CHONa
45
43
44
PhCH~OCH~ OCH
PhCHzOCHJ
45
0
b O C H 3
-----f
-bO
PhCH,OCH,
3
+
HO OH 47
46
48
Scheme 22
Another synthesis of a sugar derivative from an amino acid (L-aspartic acid), and a synthesis involving both pyruvic acid and glycine, are discussed in Section 7A. 5. THE DIELS-ALDER REACTION
The Diels-Alder reaction has been employed by a number of workers for the preparation of dihydropyrans as substrates for the synthesis of a wide
5. The Diels-Alder Reaction
25
range of monosaccharides. These examples are discussed in Section 6, which is specifically concerned with the synthesis of sugars from pyran derivatives. In this section, the use of Diels-Alder condensations in two very elegant, total syntheses of novel carbohydrates is described. The first example is that of Belleau and A u - Y o ~ n g ,whose ~ ~ objective was the total synthesis of amino sugars (Section 7A). They utilized the dienophilic properties of I-chloro-1-nitrosocyclohexane and condensed it with methyl sorbate to yield cis-3-methyl-6-methoxycarbonyl-3,6-dihydro1,Zoxazine hydrochloride (49). This Diels-Alder adduct is formed in a stereospecific manner, the cis-adducts only being formed when the diene has the trans, transgeometry, as is present in methyl sorbate. The adduct possesses a double bond at positions 4 and 5 and may therefore be hydroxylated to yield, after ring cleavage, 5-amino-5,6-dideoxy-~~-hexonic acids. The possible formation of a 2-amino-2-deoxy derivative was eliminated when the adduct, after hydrogenation and ring-opening, was shown to be an a-hydroxy acid and not an a-amino acid. When the N-benzoate of the adduct 50 was hydroxylated with osmium tetroxide-pyridine complex, attack of the reagent occurred from the least hindered side, and the diol-N-benzoyl-acid51 resulted. Mild hydrolysis of 51, followed by catalytic hydrogenation over Adam’s acid (52) (see Scheme 23). catalyst, furnished 5-amino-5,6-dideoxy-~~-allonic
49
50
1
CO,H
I
HCOH BZ = PhC-
1I
0
I
HCOH
I HCOH I
t
Bz
HO” CH 3
HCNHz
I
CH, 52
Scheme 23
51
CH, 50
54
53
d I
55
57
58
Scheme 24
26
5. The Diets-Alder Reaction 57
27
58
L
I
COzH I
6H3
H~NH, I CH,
59
Scheme 24 (contd.)
60
The double bond in the N-benzoyl adduct 50 could be converted to the epoxide by reaction with peroxytrifluoroacetic acid. The reaction was not stereospecific; both possible isomers, 53 and 54, were produced in equal amount and were separable by crystallization. Both epoxides, on reaction with formic acid, yielded a mixture of two trans-diol monoformates 55 and 56. On treatment of this mixture with methanolic hydrogen chloride, a mixture of products 57 and 58 was obtained. Mild hydrolysis of this mixture gave a single tetrahydro-I ,Zoxazine carboxylic acid (59), whereas catalytic hydrogenation of the mixture afforded 5-amino-5,6-dideoxy-~~-gulonic acid (60) (see Scheme 24). In the early 1960s there was considerable interest in the synthesis of sugars in which the ring oxygen was replaced by other heteroatoms such as nitrogen or sulfur.'ja All of the syntheses were achieved by chemical modification of readily available monosaccharides. Very recently, Vyas and Hay5Qfound that methyl cyanodithioformate possesses a very marked dienophilic activity and affords a facile one-step synthesis to a variety of unsaturated, deoxy, I-thio sugars with sulfur in the ring by way of a Diels-Alder reaction. Thus with 1,3-butadiene, methyl cyanodithioformate affords a 75 % yield of a crystalline As-dihydrothiopyran derivative 61* after 24 hours in methylene chloride at room temperature. The adduct is a stable low-melting racemate:
//
S
CII / \
CN
CHZ
+ HC I SCH,
--t
HC
\\
UCN SCH3
CH2
61
* Compound 61 is considered to be a carbohydrate derivative by virtue of the fact that it is an acetal, specifically, a dithioacetal. Compounds such as 61, which have an S-alkyl (or
S-aryl) group at C-1 of the sugar ring, are called 1-thioglycosides(see also Section 6).
28
The Total Synthesis of Carbohydrates
2,3-Dimethyl-lY3-butadieneis more reactive and reacts with methyl cyanodithioformate in 5 min at 4" to give a racemic mixture of 62 and 63:
CH,
CH 62
63
With cyclopentadiene, the reaction occurs instantaneously at 0" in methylene chloride to give an isomeric mixture of 64 and 65 in a 60:40 ratio, respectively. Isomer 64 crystallized, leaving a syrupy residue of 64 and 65. The structure
SCH 3
65
64
CN
of 3-endo-thiomethyl-3-exo-cyano-2-thiobicyclo[2.2.l]hept-5ene was assigned to 64 by nuclear magnetic resonance (NMR) spectroscopy. Isomerization of 64 into 65 was observed when 64 was kept in chloroform solution at room temperature. A stable crystalline adduct 66 has also been obtained from the reaction of frans,fruns-l,4-diacetoxy-l,3-butadiene with methyl cyanodithioformate in refluxing chloroform : OAc
OAc
SCH,
66
6. SYNTHESES FROM FURAN
AND PYRAN DERIVATIVES
It is well known that 1,4- and 1,5-hydroxyaldehydes exist primarily as cyclic hemiacetals?O
' =
CH,CH2CH2
HOI
HC=O
'L.2
Q-4" 88.6 %
93.3 %
67
68
6. Syntheses from Furan and Pyran Derivatives
29
The two hemiacetals 67 and 68 may be regarded as carbohydrate models, namely, 2,3-dideoxytetrofuranose and 2,3,4-trideoxypentopyranose,respectively. A discussion of the compositions of the equilibrium mixtures of sugars in solution has been presented recently by AngyaLB1Treatment of a cyclic hemiacetal with an alcohol in the presence of anhydrous acids yields an acetal. In the language of carbohydrate chemistry, the acetal is called a glycoside, and the conversion is said to involve the introduction of an aglycon at the anomeric center of the sugar: A iioiiieI ic
QOH
&
J+
Sugar (glycose)
Y A p l ycon
Glycoside
It is not surprising that furan and pyran derivatives themselves have been found to be useful substrates for the total synthesis of sugars. Several syntheses of simple models of sugars have been achieved by addition reactions to the double bonds in 2,3-dihydrofuran (69) and 3,4dihydro-2H-pyran (70) and their derivatives.* Compound 69 was first made,
69
70
in 24 % yield, by passing tetrahydrofurfuryl alcohol over a copper-nickel alloy;ea higher yields have been obtainedes by treatment of 3-chloro-2alkoxytetrahydrofurans (made from tetrahydrofuran) with sodium. In another method,e4 butane-l,4-diol is dehydrogenated over a cobalt catalyst at 220" to give 2-hydroxytetrahydrofuran, which then eliminates water to afford 2,3-dihydrofuran in 80 % yield. A convenient preparation is the rearrangement of the commercially available 2,5-dihydro isomer.65 3,4-Dihydro-2H-pyran (70) is made commercially by passing tetrahydrofurfuryl alcohol over alumina at 350°.e6The mechanism of this ring-expansion has been followed with 14C and found to proceed through the carbonium ion 71?'
71
*
70
The numbering of furans and pyrans follows the universally adopted system: the heteroatom is called No. 1. In carbohydrate nomenclature, the anomeric center has been given this number. Compounds such as 69 and 70 in carbohydrate chemistry are called glycals.
30
The Total Synthesis of Carbohydrates
Some a,/?-unsaturated carbonyl compounds dimerize to 3,4-dihydro-2Hpyrans;88 for example, acrolein, in the presence of a little hydroquinone, dimerizes to 2-formyl-3 ,4-dihydro-2H-pyran (72), and crotonaldehyde gives 73 :
73
72
Several substituted 3,4-dihydro-2H-pyrans have been obtained by a DielsAlder reaction, and these have served as substrates for the synthesis of sugars or were already simple models of sugars. Smith et aLBghave added a wide variety of ethylenic compounds, such as vinyl ethers, unsaturated esters, methacrylonitrile, and olefins, to acrolein and other conjugated carbonyl compounds, such as methacrolein and crotonaldehyde:
CHR"
H R"
R'C
I
RC
CXY
0
H X
2,3-Dihydrofurans and 3,4-dihydro-2H-pyrans, which have the double bond in the a-position to the ring oxygen, undergo the usual reactions of vinyl ethers. Thus they react vigorously with water and other hydroxylic compounds in the presence of a trace of acid to give mainly 2-hydroxy- or 2-alkoxy-tetrahydro compounds or esters:
The acetal (or glycoside) shown in the preceding equation is unstable in the presence of aqueous acids, giving the alcohol and 2-hydroxytetrahydropyran (or free sugar). Many simple carbohydrates have been prepared by such addition reactions, and the compounds have been used as models for studies of glycoside hydrolysis70or for heterocyclic conformational analy~is.~l Of particular importance in the preparation of 2-oxy-substituted tetrahydro(or Edward-Lemieux effect79),by which term pyrans is the anomeric
6. Syntheses from Furan and Pyran Derivatives
31
is meant the greater preference of an electron-withdrawing group for the axial position when it is located adjacent to a heteroatom in a ring than when it is located elsewhere. Thus, for example, it has been found" that acidcatalyzed addition of methanol to 2-methoxyrnethyl-2,3-dihydro-rlH-pyran gave an equilibrium mixture of two isomers in the ratio of 70% trans to 30% cis (Scheme 25). Various 2-tetrahydrofuranyl ethers (or furanosides) have been prepared76 by the addition of alcohols to 2,3-dihydrofuran in the presence of acid. An alternative synthesis, from tetrahydrofuran and t-butyl perbenzoate in the presence of alcohols, is also a~ailable.?~ 2,3-Dihydroxy-tetrahydropyranand -tetrahydrofuran can be made from the dihydropyran 70 and the dihydrofuran 69 by reaction with osmium CH20CH3
H2C
OCH3
+ C H , O H -@+
H %
OCH, trans (70%)
+%
CH2OCH3
OCH3
H cis (30%)
Scheme 25
tetroxide and hydrogen peroxide in t-butan0177 or with lead tetraacetate,BB respectively. Both diols give 2,4-dinitrophenylosazones. In a more recent study,78 70 was treated with m-chloroperoxybenzoic acid in wet ether to give a diol (Scheme 26) whose NMR spectrum indicated it to be a mixture of the cis and trans isomers in the ratio of 30: 70, respectively. The same ratio was obtained when the diol mixture was allowed to equilibrate in the presence of a small amount of p-toluenesulfonic acid. It therefore appears that if the reaction of the dihydropyran with the peroxy acid had taken place by way of an epoxy intermediate and produced initially the trans-diol stereospecifically, a rapid isomerization must have occurred to form the equilibrium mixture. Treatment of the crude diol with @-chloroethanol in the presence of p-toluenesulfonic acid produced 2-(~-chloroethoxy)-3-hydroxytetrahydropyran (74) as a mixture of the cis and trans isomers in a ratio of 69 : 31. Upon further treatment of the mixture with a catalytic amount ofp-toluenesulfonic acid in ,4-chloroethanol, this ratio changed to the equilibrium mixture of 40% cis: 60% trans. Treatment of 74 with sodium hydride in 1 ,Zdimethoxyethane afforded both the cis and trans isomers of tetrahydropyrano[2,3-b]-1 ,I-dioxane. The results obtained with 74 indicate that the reaction of the diol with /3-chloroethanol is highly stereospecific. A possible explanation7B for this stereospecificity involves hydrogen bonding by the C-3
32
The Total Synthesis of Carbohydrates
/
DM E
Scheme 26
hydroxyl group with the incoming alcohol, and hence a preference for attack by the 6-chloroethanol (by way of an SN2 or S,1 mechanism) on the same side of the ring as is occupied by the C-3 hydroxyl. Partial isomerization would then yield the 69:31 ratio of cis to trans isomers (Scheme 27). In a related study, Sweet and Brown7e performed the oxidation of 2,3dihydrofuran and 3,4-dihydro-2H-pyran with peroxybenzoic acid or mchloroperoxybenzoic acid in the presence of an alcohol and obtained, respectively, trans-2-a1koxy-3-hydroxytetrahydrofurans and trans-2-alkoxy3-hydroxytetrahydropyrans. Presumably, oxidation with the peroxy acids involves the formation of an epoxide intermediate or, alternatively, an “epoxidelike” transition state; the observed results are then in agreement with the known preferred trans opening of an epoxide ring. In the presence of acids, the acetals isomerized readily to given an equilibrium mixture of cis and trans isomers. The reactions of 3,4-dihydro-2H-pyran (70) with halogens and halogen compounds have been well studied, and the products have proved to be useful intermediates for further syntheses. Compound 70 readily adds chlorine, bromine, hydrogen chloride, or hydrogen bromides0 to give 2,3-dihalogeno- or 2-halogeno-tetrahydropyrans.The halogen atom at C-2 is removed as hydrogen halide, on distillation of the product at atmospheric pressure, to give 3-chloro- or 3-bromo-5,6-dihydro-4H-pyran(75, Scheme 28). The halogen atom at C-3 in all of these compounds is relatively inert, but that at C-2 resembles the halogen in a-chloro ethers. With alcohols or sodium salts of aliphatic acids, the dihalogeno compounds give 2-alkoxy or 2-acyloxy compounds, and with water substituted bis(tetrahydropyrany1) ethers are obtained.81*82Reaction of compound 70 or its derivatives with halogens in a hydroxylic solvent also gives the corresponding halogenated 2-hydroxy- or 2-alkoxy-tetrahydropyran;for example, 3-chloro-2-hydroxytetrahydropyran (76) can be made by chlorinating an emulsion of compound
D@
0
a
“ I I
0
z ,
p z
0
M
z
33
34
The Total Synthesis of Carbohydrates
70 in water. The mechanisms of halogenation and halogenomethoxylation
of compound 70 have been discussed by Lemieux and F r a ~ e r - R e i d . ~ ~ Addition of hydrogen chloride to 2,3-dihydrofuran (69) affords 2-chlorote t r ah y d r o f ~r a n.~~ 2,3-Dichlorotetrahydrofurancan be made by chlorinating compound 69 or, more conveniently, tetrahydrofuran itself. The chlorine atom at C-2 is as reactive as that in 2-~hlorotetrahydropyran,and a corresponding series of hydroxy- and alkoxy-chlorotetrahydrofurans can be prepared.63***
O ' ('-JR 70
76
75
Scheme 28
The preparation of 3-alkythio-2-methoxy-tetrahydrofurans and -tetrahydropyrans has also been achieved. Senning and Lawessone5have described the synthesis of several 2-alkoxy-3-(trichloromethylthio)tetrahydrofurans by what was considered to be an "acid-catalyzed" addition of trichloromethylsulfenyl chloride to 2,3-dihydrofuran, followed by displacement of the halogen by an alkoxy group. More recently, Baldwin and Brown reported the results obtained when a similar procedure was applied to 3,4-dihydro2H-pyranas and examined in detail the mechanism of the addition of ethanesulfenyl ~ h l o r i d e . I~t?was foundee that the reaction at -20" of alkylsulfenyl chlorides with 3,4-dihydro-2II-pyran, followed by treatment of the product with sodium methoxide, gave, highly stereoselectively, trans-3-alky lt hio-2methoxytetrahydropyrans. Distillation under vacuum a: temperatures above 55" resulted in the elimination of methyl alcohol to produce 5-alkylthio-3,4dihydro-2H-pyrans in good yield. At lower temperatures, isomerization to a mixture of cis- and 1rans-3-alkylthio-2-methoxytetrahydropyransoccurred. lsomerization also occurred with acid catalysts. A mechanism for the formation of the 2-alkoxy-3-alkylthiotetrahydropyrans was p r o p o ~ e d ~ ~ ~ ~ ~ and is shown in Scheme 29. The first step involved a nucleophilic attack of the a$-unsaturated ether on the alkylsulfenyl chloride to displace the
6. Syntheses from Furan and Pyran Derivatives
RS
35
R = CH,, CaHL, PhCHt Scheme 29
chloride ion and give an oxocarbonium-episulfonium ion in which the episulfonium contribution is considered to be paramount. Attack by chloride ion would then give, reversibly, truns-3-alkylthio-2-chlorotetrahydropyran. In the presence of an alcohol, the episulfonium ion, obtained by loss of the chloride ion, then gave, stereoselectively, the frans-3-alkylthio-2-alkoxytetrahydropyran. In recent years, there has been considerable interest in the utility of 2-alkoxy-5,6-dihydro-2H-pyrans (77) and their derivatives as substrates for the synthesis of a wide range of monosaccharides. In addition to their use for the preparation of simple model carbohydrates, they have been intermediates in the total synthesis of some sugar moieties found in antibiotics (discussed later in this section). The 2-alkoxy-5,6-dihydro-2H-pyranscan be prepared by conversion of 3 ,Cdihydro-2H-pyran (70) into 2-alkoxy-3halogenotetrahydropyrans followed by dehydrohalogenation.
36
The Total Synlhesis of Carbohydrates
Sweet and Brownaa have investigated the epoxidation of some 2-alkoxy5,6-dihydro-2H-pyrans and the lithium aluminum hydride reduction of the resultant epoxides. Thus 2-methoxy- and 2-r-butoxy-5,6-dihydro-2H-pyran were found to react with rn-chloroperoxybenzoic acid in ether solution, at 35", to give mixtures of trans- and cis-2-alkoxy-3,4-epoxytetrahydropyrans in which the trans to cis ratio was 3: 1 and 9 : 1 for the 2-methoxy and 2-rbutoxy compounds, respectively (Scheme 30). This selectivity can clearly be attributed to the steric effect of the 2-alkoxy group. Lithium aluminum hydride attacks the epoxide ring in these cis and trans compounds exclusively at the epoxide carbon remote from the alkoxy substituent to form only 2-alkoxy-3-hydroxytctrahydropyrans. This selectivity of hydride attack was
R R
=
CH,
= C(CH,),
25 % -10%
75 %
-90 %
Scheme 30
attributed to the polar effect of the two oxygen atoms at the anomeric carbon (C-2) in the tetrahydropyrans. The foregoing approach was later employed by Sweet and Browns9 to prepare a mixture of methyl 4-deoxy-3-0-methyl-a- and /3-~~-rkreo-pentopyranosides starting with the cis-, trans-, or cis-rrans-2-alkoxy-3,4-epoxytetrahydropyran (78 and 79). The mixture was obtained by a highly selective reaction of methanol with 78 and/or 79 in the presence of a catalytic amount of p-toluenesulfonic acid (Scheme 31). The result can be rationalized on the assumption that the first step was attack by methanol of the protonated epoxide exclusively at the carbon atom remote from the anomeric center. This was then followed by a slower isomerization to produce an equilibrium mixture of the a and anomers of the DL mixture. Sweet and Brownso have also found that acid-catalyzed methanolysis of 2-methoxy-5,6-dihydro-2H-pyrangave, in good yield, a 4:1 mixture of trurrs- and cis-2,4-dimethoxytetrahydropyran,respectively. The two isomers could be readily separated by gas-liquid chromatography. The formation essentially of only the two components shows that a highly selective reaction had occurred between methanol and 2-methoxy-5,6-dihydro-2H-pyran.A possible route is shown in Scheme 32. Because of the electron-withdrawing effect of the two oxygen atoms attached to the anomeric carbon, the more stable protonated state is that in which the carbon of the double bond more
.1
m.CIC,H,CO,H
78
79
H I
OCH,
79
I
H@,
[email protected],OH
0 tI
e
o
C
H
3
OH Scheme 31
37
38
The Total Synthesis of Carbohydrates
remote from the anomeric center becomes positive. Another route to the formation of the 2,4-dimethoxytetrahydropyranshas also been considered by Sweet and Brown; this involves a preferred protonation of the oxygen atoms rather than the double bond; acetals are known to be unstable in an acidic medium. Treatment of a solution of 2,4-dimethoxytetrahydropyran, in a mixture of water and 1,2-dimethoxyethane with Amberlite IR-120 gave 2-hydroxy-4-met hoxytetrahydropyran as an equilibrium mixture of cis and tram isomers in the ratio 1 : l.@l The reactions of 2-methoxy-5,6-dihydro-2H-pyranwith 1,3-dibromo-5,5-
80
81
Scheme 33
dimethylhydantoin in ether-methanol,el and with ethanesulfenyl chloride,ea have been described by Baldwin and Brown. The former reaction gave a 2: 1 mixture of the isomers 3p-bromo-2a,4a-dimethoxytetrahydropyran(80) and 3a-bromo-2a,4~-dimethoxytetrahydropyran (81), respectively. The structures and preferred conformations of the isomers are shown in Scheme 33. The reaction of ethanesulfenyl chloride with 2-methoxy-5,6-dihydro-2H-pyran gives only 4/?-chloro-3ol-ethylthio-2/?-methoxytetrahydropyran(82). A proposede2 route for this highly selective reaction is shown in Scheme 34. Because of the anomeric effect, the preferred conformer of 2-methoxy-5,6dihydro-2H-pyran is considered to be that in which the C-2 methoxy group is quasi axial. The first step presumably is electrophilic attack of ethanesulfenyl chloride on the dihydropyran from thc least hindered side of the molecule, namely, trans to the C-2 methoxy group, to give an episulfonium ion intermediate. The next step is attack by the chloride ion at either C-3 or C-4 with simultaneous opening of the episulfonium ring. It is believed that the steric effect of the quasi axial methoxy group, and its polar repulsion for the chloride ion, would strongly inhibit attack by the chloride ion at C-3. Moreover, the electron-withdrawing effect of the two oxygen atoms at (2-2, which would destabilize an incipient positive charge at C-4 less than C-3, favors reaction of the chloride ion at C-4 rather than at C-3. The reaction of bromine with 2-ethoxy-5,6-dihydro-2H-pyranhas also been studiedezn; two isomeric dibromides were obtained, namely, 3a,4,!3-dibromo-2a-ethoxytetrahydropyran and 3a,4/l-dibromo-2/?-ethoxytetrahydropyran.Treatment of
6. Syntheses from Furan and Pyran Derivatives
39
I
CsHr,
I
C2H5S
I CZHa
82
Scheme 34
these isomers with refluxing ethanolic sodium ethoxide afforded trans5,6-diethoxy-5,6-dihydro-2H-pyran, cis-2,5-diethoxy-5,6-dihydro-2H-pyran, rrans-2,5-diethoxy-5,6-dihydro-2H-pyran, and 3-bromo-2-ethoxy-5,6dihydr0-2H-pyran.~~" Cahu and Descotess3 have also subjected 2-alkoxy-5,6-dihydro-2H-pyran to addition reactions such as hydrohalogenation, hydroxyhalogenation, epoxidation, and cis-hydroxylation, and the cis- and trans-2-alkoxy-3,4epoxytetrahydropyrans, to acid hydrolysis, and addition of d i methylamine. The potential scope of additions to 2-alkoxy-5,6-dihydro-2H-pyrans has recently been extended by the availability of some new derivatives. It has been s h o ~ n ~that * * the ~ ~condensation of esters of glyoxylics@or mesoxalic acid with l-alkoxy-l,3-butadienes affords esters of 2-alkoxy-5,6-dihydro2H-pyran 6-carboxylic acid (83) and 2-alkoxy-5,6-dihydro-2H-pyran 6,6respectively, in good yield. The butyl esters of 83 dicarboxylic acid (M), have been converted with lithium aluminum hydride into 2-alkoxy-5,6dihydro-6-hydroxymethyl-2H-pyrans which, on hydrogenation over platinum, gave the corresponding 2-alkoxy-6-hydroxymethyltetrahydropyrans.g7 Treatment of some of the 6-substituted 2-alkoxy-5,6-dihydro-2H-pyrans with m-chloroperoxybenzoic acid yielded mixtures of stereoisomeric epoxides, R = Me, Et, Pr, Bu R" = E t , B u R"' = H 84 R" = E t
83
R"'= COOEt
40
The Total Synthesis of Carbohydrates
which were separated by column c h r o m a t ~ g r a p h y .Methanol ~~ has also been added, under acidic conditions, to 6-substituted 2-methoxy-5,6dihydro-2H-pyrans to yield derivatives of 2,4-dimethoxytetrahydropyran.0e The addition of bromine in ether solution to 6-substituted 2-methoxy-5,6dihydro-2H-pyrans leads in the case of trans isomers to two stereoisomeric dibromo derivatives having the u-.xy/oand u-arubino configurations, whereas, cis isomers give #?-arabino dibromo derivatives; in methanol solution, the treatment with bromine yields mixtures of dibromo and bromomethoxy derivative^.^^" Stereospecific syntheses of four diastereomeric methyl 4deoxy-DL-hexopyranosides have also been achieved from esters of 2-methoxy5,6-dihydro-2H-pyran 6-carboxylic acid.OobZwierzchowska and Zamojski’oo have demonstrated a “glycal-pseudoglycal” rearrangement101 with some &substituted dihydropyran derivatives. Thus, for example, heating trans-85 (a product of the condensation of l-acetoxy-l,3-butadiene with butyl glyoxylate) in acetic anhydride gave two new products, the butyl esters of trans- and cis-4-acetoxy-5,6-dihydro-4H-pyran 6-carboxylic acid (86). The isomerization could be reversed, when cis- or truns-86 were heated in acetic anhydride: OAc
85
86
Much progress has been made in recent years by Brown and his associates
in Canada toward the total syntheses, involving pyran intermediates, of
monosaccharides. Some of this work has already been presented in this section. Other syntheses are now described which incorporate many of the reactions already discussed. The synthesis of methyl 2,3-anhydro-4-deoxy-6-O-methyl-a-~~-lyxohexopyranoside (91) has been achievedlO* from 2-methoxymethyl-3,4dihydro-2Wpyran (87). The sequence of reactions for this synthesis is shown in Scheme 35. Bromomethoxylation of 87 gave a 9 : l mixture of the two isomers 88 and 89, respectively. This mixture was heated with sodium methoxide in refluxing methanol to give a product that contained at least 95 % of the trans-Zmethoxy-6-methoxymethyl-5,6-dihydro-2H-pyran, 90. rn-Chloroperoxybenzoic acid converted 90 into a mixture of methyl 2,3anhydro-4-deoxy-6-O-methyl-a-~~-lyx-o-hexopyranoside (91) (>95 %) and methyl 2,3-anhydro-4-deoxy-6-O-methyl-a-~~-ribo-hexopyranoside (92) 3-(%.)
6. Syntheses from Furan and Pyran Derivatives
87
88
92
91
41
89
90
Scheme 35
Earlier in this section the preparation of cis- and rrans-2,4-dimethoxytetrahydropyran was described. Recently, the synthesis of cis- and truns-2,5dimethoxytetrahydropyran and the homolog cis- and rrans-2,5-dimethoxy6-methyltetrahydropyran were reported.lo3 The starting material for the synthesis of cis- and rrans-2,5-dimethoxytetrahydropyran (see Scheme 36) was 2 methoxy-3,4 dihydro-2H pyran (93), the condensation product of acrolein and methyl vinyl ether. Hydroboration and subsequent oxidation of 93 gave a 1 :2 mixture of cis- and trans-5-hydroxy-2 methoxytetrahydropyran (94). From the methylated product 95, trans-2,5-dimethoxytetrahydropyran could be obtained by gas-liquid chromatography. Distillation of the mixture of cis- and trans-2,5-dimethoxytetrahydropyran(95) over a catalytic amount of phosphoric pentoxide gave 3-methoxy-3,4-dihydro-2Hpyran (96). Bromomethoxylation of 96 with 1,3-dibrom0-5,5-dimethylhydantoin in ether-methanol gave a mixture of three isomers, from which the major product, 97, could be separated. Hydrogenation of 97 over palladium-on-charcoal afforded cis-2,5-dimethoxytetrahydropyran(98). Both cis- and trans-2,5-dimethoxy-6-methyltetrahydropyran(100 and 101) were obtained by hydroboration and oxidation of 2-methoxy-6-methyl3,4-dihydro-2H-pyran (99) (the condensation product of methyl vinyl ketone and methyl vinyl ether), followed by methylation of the resultant product (as shown on page 42).
The Total Synthesis of Carbohydrates
42
95
94
93
cH30w OC H
OCH 3
91
98
Scheme 36
Srivastava and Brownloq have utilized 3a-bromo-2a,5a-dimethoxytetrahydropyran (99, obtained by bromomethoxylation of 3-methoxy-3,4dihydro-2H-pyran (96) as shown in Scheme 36, to prepare methyl 4-0methyl-a-DL-arabinopyranoside(104) (see Scheme 37). Compound 97 was first dehydrobrominated with a boiling solution of potassium hydroxide in methanol to give cis-2,5-dimethoxy-5,6-dihydro-2H-pyran (102). Treatment of 102 with m-chloroperoxybenzoic acid in methylene chloride afforded, almost exclusively, the epoxide 103, which was converted into methyl 4-O-methyl-a-~~-arabinopyranoside (104) with aqueous potassium hydroxide.
99
6. Syntheses from Furan and Pyran Derivatives
96 -+
97
43
OCH,
--*
103
104 Scheme 37
A particularly significant development, for the total synthesis of monosaccharides, has been the conversion of acrolein dimer (72) into the olefins 6,8-dioxabicyclo[3.2.l]oct-3-ene (109) and 6,8-dioxabicyclo[3.2.l]oct-2-ene (110) by the sequence of reactions shown in Scheme 38.’O5.’O6 Acrolein dimer was first reduced with sodium borohydride to give 2-hydroxymethyl-3,4dihydro-2H-pyran (105). When the alcohol 105 was heated in refluxing benzene containing a catalytic amount of p-toluenesulfonic acid, 6 3 dioxabicyclo[3.2.l]octane (106) was formed. Treatment of 106 with bromine in carbon tetrachloride gave a mixture of two isomeric monobromides, considered to be trans- and cis-4-bromo-6,8-dioxabicyclo[3.2.l]octane (107
72
109
Scheme 38
I10
44
The Total Synthesis of Carbohydrates
and 108). Heating the mixture of 107 and 108 in refluxing ethanolic potassium hydroxide gave the two olefins 109 and 110, which were readily separated by gas-liquid chromatography; the proportion of the two isomeric olefins obtained depended upon the proportion of base to monobromide used during dehydrohalogenation. 6,8-Dioxabicyclo[3.2.l]oct-3-ene(109) and its isomer 110 have been used for the synthesis of several monosaccharides and their derivatives. Thus, for example, 109 has been converted into l16-anhydro-4-deoxy-#?-~~-xylohexopyranose (113) by epoxidation with rn-chloroperoxybenzoic acid to give ribo-hexopyranose (111) ,contaminated with 1,6 :2,3-dian hydro4-deoxy-P-~~< 5 % of the isomeric 1,6:2,3-dianhydro-4-deoxy-#?-~~-lyxo-hexopyranose (112), followed by treatment with aqueous potassium hydroxide (Scheme 39).105 The stereoselective synthesis of DL-glucose has been ac~omplished107~~~7n in 34 % overall yield starting from 1,6:2,3-dianhydro-4-deoxy-B-~~-ribo-hexopyranose (111) (Scheme 40).Compound 111was converted into 1,6-anhydro3,4-dideoxy-~-~~-eryrkro-hex-3-enopyranose (114, R = H) with n-butyllithium. Treatment of 114 with rn-chloroperoxybenzoic acid gave 1,6: 3,4dianhydro-p-m-ah-hexopyranose (119, which with aqueous barium
112
111
OH 113
Scheme 39
6. Syntheses from Furan and Pyran Derivatives
114
45
115
117
116
Scheme 40
hydroxide was converted into 1,6-anhydro-/?-~~-gluco-hexopyranose (116). Acid-catalyzed hydrolysis of 116 afforded a,@-DL-glucose(117). 3-@Methylhave also been obtained107a m-glucose and 3-deoxy-~~-ribo-hexopyranose from compound 115. Thus, treatment of the oxirane 115 with sodium methoxide in methanol gave a very good yield of 1,6-anhydro-3-0-methyl-@ DL-gluco-hexopyranose as the only isolable product. Acid-catalyzed hydrolysis of the diacetate of this product afforded an excellent yield of 3-0-methyl-a,/?-~~-glucopyranose. 1,6-Anhydro-3-0-methyL/?-~~gluco-hexopyranose could also be obtained, along with methyl 3-0-methyla,p-DL-glucopyranoside, when 115 was heated in methanol containing p-toluenesulfonic acid monohydrate. Lithium aluminum hydride reacted with 115 to form 1,6-anhydro-3-deoxy-~-~~-ribo-hexopyranose, which was hydrolyzed readily by acid to 3-deoxy-~~-ribo-hexopyranose. In the above work of Singh and B ~ o w all~ of ,the~ products ~ ~ obtained ~ ~ ~from ~ the ~ oxirane 115 were those resulting only from trans diaxial opening of the oxirane ring. Total syntheses of a,@L-allose (118) and a,p-DL-galactose (119) and their 2-0-methyl derivatives have also been accomplished*08 from 114 (R = H, CH,- or CH,CO-) by way of a stereoselective cis-hydroxylation by osmium tetroxide. The observation that the proportion of allo to galacto product obtained varied with the solvent (pyridine or dioxane) used in the hydroxylation reaction was interpreted as showing that the attack by osmic acid is subject to steric approach control. Very recently, the total synthesis of several monodeoxy- and dideoxy-DLhexopyranoses from the olefins 109 and 110 was reported.loSThe reactions
46
The Total Synthesis of Carbohydrates
HO
HO 118
119
performed with olefin 109 are shown in Scheme 41. Reaction of osmic acid (120), which, with 109 gave 1,6-anhydro-4-deoxy-fi-~~-ribo-hexopyranose on acid-catalyzed hydrolysis in aqueous dioxane, afforded 4-deoxy-a,@-~~do-hexopyranose (121). Conversion of 109 into 1,6: 2,3-dianhydro-4deoxy-#?-DL-ribo-hexopyranose (ill), followed by treatment of 111 with lithium aluminum hydride, gave 1,6-anhydr0-3,4-dideoxy-@-~~-erytkrohexopyranose (122). Compound 122 was hydrolyzed to 3,4-dideoxy-a,/?-~~erythro-hexopyranose (123). With olefin 110, the same sequence of reactions led to 2-deoxy-a~-~~-ribo-hexopyranose and, presumably, 2,3-dideoxyu,#?-DL-eryt/iro-hexose, although the latter compound was not Fully characterized. Other examples of the stereoselective hydroxylation of double bonds by osmic acid were provided by the total synthesis of 4-O-methyl-~~-lyxose and 4-deoxy-~~-ribose~~O (see Scheme 42). The reaction of osmium tetroxide in py r id ine with 2-met hoxy-5,6- di hydro-2H-pyran and cis-2,5-d imethoxy5,6-dihydro-2H-pyran afforded in excellent yield, as the sole isolable products, methyl I-deoxy-f?-~~-erythro-pentopyranoside (124) and methyl
0
-
#
oso,
ni-CIC, H,COIH
dioxane or pyridine
111 0
109
1
1
IN HCI,
LiAIH,
@ 122 OtI
120
dloxanc
INHCI
CH,OH
dioxanc
123
Scheme 41
HRo CH,OH
121
6.
Syntheses from Furan and Pyran Derivatives
OCH,
It R=H R = CH,O
47
OCH
124 R = H 125 R = CH,O
126 R = H 127 R = CH,O
Scheme 42
4-O-methyl-a-~~-lyxopyranoside (125), respectively. Hydrolysis of 124 and
125 with aqueous dioxane containing sulfuric acid gave 4-deoxy-~~-eryfhropentose (4-deoxy-~~-ribose,126) and 4-O-methyl-~~-lyxose(127). The attack by osmic acid occurs nearly completely, if not exclusively, from the unhindered side, remote from the substituent. These results indicate that 2-methoxy-5,6-dihydro-2H-pyran must be predominantly in the conformation in which the C-2 alkoxy group is quasi axial. Many antibiotics have been found to contain sugars of unusual structure, such as aminodeoxy, deoxy, and branched-chain sugars. During the 1960s, total syntheses, involving pyran intermediates of some of these carbohydrate moieties have been achieved. In 1962 Korte et al."' described a synthesis of DL-desosamine (DL-picrocin). Desosamine is a component of several rnacrolide antibiotics including erythromycin, oleandomycin, and narbomycin; its structure has been shown to be that of 3,4,6-trideoxy-3-dimethylamino-~-xylo-hexose (128). The starting material (see Scheme 43), b-caprolactone (129), was reduced with lithium aluminum hydride in tetrahydrofuran to give 2-hydroxy-6-methyltetrahydropyran(130), which, on being heated in the presence of alumina, afforded 5,6-dihydro-6-methyl-4Hpyran (131). Compound 131 was then converted into 2,3-dibromo-6methyltetrahydropyran (132) with bromine in carbon tetrachloride. Treatment
128
48
The Total Synthesis of Carbohydrates
of 132 with ethanol saturated with ammonia gave the glycoside, 3-bromo-2ethoxy-6-methyltetrahydropyran (133), which, when refluxed in ethanol containing sodium, yielded 5,6-dihydro-2-ethoxy-6-methyl-2H-pyran(134). Oxidation of 134 with peroxybenzoic acid afforded 3,4-epoxy-2-ethoxy-6methyltetrahydropyran (135). Opening of the epoxide ring with aqueous dimethylamine gave 4-dimethylamino-2-ethoxy-3-hydroxy-6-methyltetra-
129
134
135
130
133
131
132
136
DL-Desosamine
Scheme 43
hydropyran (136). Hydrolysis of 136 with aqueous hydrochloric acid finally afforded DL-desosamine hydrochloride. Newman112has also achieved a synthesis of DL-desosamine by way of 5,6-dihydro-2-ethoxy-6-methyl-2H-pyran(134). This preparation of 134 is shown in Scheme 44. The lithium salt of propargyl aldehyde diethyl acetal reacted with propylene oxide to give I , I-diethoxy-5-hydroxyhex-2-ynein 60% yield. This compound was converted into 134 by hydrogenation over 10% palladium-on-charcoal, followed by treatment with acid. More recently, Mochalin et al.lI3 have reported a stereospecific synthesis of desosaniine by way of the reaction of 135a with dimethylamine. Yasuda and Matsumoto have recently described total syntheses of some sugars found in antibiotics. One of these syntheses was that of methyl DL-mycaminoside (137).lI4 Mycaminose is a sugar component in the antibiotics magnamycin, spiramycin, and leucomycin. Hydroboration of
6. Syntheses from Furan and Pyran Derivatives
0
/\
CH,-CH-CHz
+ HC=C-CH(OCzHJz
I
I
OH (HI. H@
H IC
49
I
5-
DuLi
CH+2H--CHZ--C~C--CH(OC,H&
134 Scheme 44
3,4-dihydro-2-ethoxy-6-methyl-2H-pyran(138), followed by oxidation with hydrogen peroxide in alkaline solution, gave the alcohol 139 (see Scheme 45). Treatment of 139 with bromine in boiling methanol containing hydrogen chloride yielded three bromo compounds, 140, 141,and 142. Each of these three compounds could be converted into the unsaturated alcohol 143; 142, for example, whose NMR spectrum suggested that the bromine and methoxyl groups are diaxial (one-proton doublet at T 5.31, J 1 Hz), readily afforded 143 on treatment with sodium azide in N,N-dimethylformamide at 120-125". Oxidation of 143 with peroxybenzoic acid gave the epoxide 144, which afforded methyl DL-mycaminoside when treated with a saturated aqueous solution of dimethylamine. Yasuda and Matsumoto have also prepared methyl DL-oleandroside (145) and its C-3 epimer, methyl DL-cymaroside (146).l16 Oleandrose is a sugar component of cardiac glycosides and of the antibiotic oleandomycin. A key intermediate in this work was the unsaturated alcohol 143,whose preparation from 3,4-dihydro-2-ethoxy-6-methyl-2H-pyranhas already been described. Compound 143 was first transformed into the benzyl ether 147 with benzyl
135a
137
50
The Total Synthesis of Carbohydrates
J
I
140
142
141
143
144
137
J
Methyl DL-rnycarninoside Scheme 45
chloride and sodium hydroxide (see Scheme 46). Treatment of 147 in refluxing methanol with a catalytic amount ofp-toluenesulfonic acid gave two adducts, 148 and 149. A similar addition of methanol to 5,6-dihydro-2-methoxy-2Hpyran has already been described in this section. Hydrogenolysis of 148 over palladium-on-carbon afforded methyl DL-oleandroside (145), whereas hydrogenolysis of 149 in methanol containing hydrogen chloride yielded an anomeric mixture of methyl DL-cymaroside (146). The diamino sugar kasugamine 150 is a component of the antibiotic
H,CO
H ,CO" 145
a:3 QCH3
OCH,
146
t
143 R = H 147 R = CHpPh
I
QCH3
H3Cd' 148
1
149
145
146
Methyl ~~-01eandroside
Methyl DL-cymaroside
Scheme 46
kasugamycin. Two syntheses of derivatives of kasugamine involving pyran intermediates have recently been achieved. The starting material for the first synthesis116 (Scheme 47) was 3,4-dihydro6-methyl-2H-pyran-2-one (151). Treatment of 151 with nitrosyl chloride in methylene chloride at -60" gave a dimer of 6-chloro-6-methyl-5-nitrosotetrahydropyran-Zone (152) in 97 % yield. This dimer was easily hydrolyzed with water to give 4-oximino-5-oxohexanoic acid (153) which, on reduction with hydrogen over platinum, afforded stereoselectively DL-erythro-4amino-5-hydroxyhexanoic acid (154). The acid 154 was lactonized by treatment with acetic anhydride at room temperature to give the N-acetylated lactone 155. Compound 155 was reduced with lithium aluminum hydride to the hemiacetal 156 which, by refluxing with acetic anhydride and pyridine, gave the N-diacetyldihydropyran 157. Treatment of 157 with nitrosyl chloride gave the chloronitroso dimer 158; displacement with methanol in the presence of mercuric cyanide afforded the a-glycoside 159. Methyl DL-kasugaminide
150
a,,, 52
The Total Synthesis of Carbohydrates
-
0
151
-,*N
HAc
0~ 0155J k H 3
1 t---
(CH2)2COOH I H~NH,
I
HCOH
I
CH 154
-----, 157
Scheme 47
(160) was finally obtained from 159 by reduction over platinum, followed by hydrolysis with barium hydroxide. The erylliro acid 154 has also been converted into the sugar m-forosamine 161 in three steps. Forosamine has been obtained from the acid hydrolysate of spiramycin. The second synthesisll' of a derivative of kasugamine involved 3,4dihydro-2-ethoxy-6-methyl-2H-pyran (138). Hydroboration of 138, followed by treatment with chloramine, gave the amine 162, which was isolated as the acetylated derivative 163 (see Scheme 48). Treatment of 163 with bromine
161
6. Syntheses from Furan and Pyran Derivatives
53
containing hydrogen chloride yielded three bromo compounds, 164, 165, and 166; both of the isomers 165 and 166 could be converted into 164. Compound 164 yielded the azide 167, on treatment with sodium azide in methyl sulfoxide at 100-105". Catalytic hydrogenation of 167 produced the amine 168, which afforded the diacetyl derivative 169. Optical resolution of
EtO" QCH3 138
-
EtO
3
162 R = H 163 R = AC
J 164
165
R
AcHN
EtO
M c0'
167 R = N, 168 R =NHa 169 R = N H A c
166
170
Scheme 48
the amine 168 was effected with D-tartaric acid. The resolved amine was acetylated to give optically active 169, which was converted into the methyl or-glycoside 170 in methanolic hydrogen chloride. The synthesis of D-170 is tantamount to the total synthesis of the antibiotic kasugamycin 171, since D-170 had already been condensed with the inositol derivative (see Section 9) to give kasuganobiosamine,"* which, i n turn, had been converted into ka~ugarnycin.~1* The final example of a synthesis involving a pyran intermediate of a sugar moiety of an antibiotic to be discussed is that of DL-chakose from acrolein dimer.120Chalcose (lankavose) is a constituent of the antibiotics chalcomycin, Iankarnycin, and neutramycin; its structure has been shown to be that of
54
The Total Synthesis of Carbohydrates
11
HOOC-C-N
H
cH3
T H # 0 H
NH
0 0tl 171
4,6-dideoxy-3-U-rnethyl-~-xy/o-hexose(172). Acrolein dirner 72 was (111) as converted into 1,6: 2,3-dianhydro-4-deoxy-~-~~-ribo-hexopyranose outlined in Schemes 38 and 39. Compound 111 was converted into 1,6anhydro-4-deoxy-3-0-rnethyl-~-~~-xylo-hexopyranose (173) by reaction with acidic methanol or sodium methoxide in methanol (see Scheme 49). When heated in refluxing methanol containing Amberlite IR-120 (H+), compound 173 gave a 2 : 1 mixture of the a and fl isomers of methyl 4-deoxy3-O-rnethyl-~~-xy/o-hexopyranoside(174). Conversion of 174 into the corresponding di-0-p-toluenesulfonates 175, followed by their selective reduction with lithium aluminum hydride, gave a mixture of the a and fl isomers of methyl 4,6-dideoxy-3-0-rnethyl-2-O-p-tolylsulfonyl-~~-xyl~hexopyranoside (176). Saponification of 176 with sodium methoxide in methanol afforded a 2 : l mixture of the a and isomers of methyl 4,6dideoxy-3-O-methyl-~~-xy/o-hexopyranoside (177) (methyl DL-chalcoside) from which the a isomer was obtained pure by column or gas-liquid chrornatography. Acid-catalyzed hydrolysis of the a isomer in aqueous dioxane finally afforded a mixture of the a and P isomers of DL-chalcose. Lukes et aI.lzL have described the preparation of 5,6-dideoxy-~~-ribohexitol (181, Scheme 50) by way of a cis-hydroxylation of 2-ethyl-5-oxo-2,5dihydrofuran (179). The starting material for this synthesis was homolevulinic acid which, on distillation with 85 % phosphoric acid, gave 2-ethyl-5-0x04,5-dihydrofuran (178); 178 was isomerized to 179 with triethylamine. Hydroxylation of 179, by treatment with osmium tetroxide and sodium acid lactone (180). Reduction hypochlorite, gave 5,6-dideoxy-~~-ribo-hexonic of 180 with lithium aluminum hydride in tetrahydrofuran afforded 181. A general approach to the total synthesis of monosaccharides from furan compounds has been developed by Achrnatowicz et a1.122nThe method is
172
i
T 3:
V
55
56
The Total Synthesis of Carbohydrates
O
178
179
=
c
I
CH,OH
I ' 1 I I
HC
HCOH
CH2
CH2
CH3
CH3
180
181
I
I
Scheme 50
outlined in Scheme 50a. A 2-furylcarbinol is converted, by treatment with bromine in methanol, into a mixture of cis and trans isomers of the corresponding 2,5-dimethoxy-2,5-dihydrofuranderivative. Mild acid hydrolysis results i n a cleavage of the acetal bonds and formation of a dicarbonyl intermediate which cyclizes to the 2,3-dideoxy-~~-2-enopyranos-4-ulose; treatment with methyl orthoformate in the presence of a Lewis acid catalyst yields a mixture of methyl glycosides which can be separated by column or gas-liquid chromatography. Reduction with metal hydrides122b leads to stereoisomeric methyl 2,3-dideoxy-~~-2-enopyranosides. The method has been applied122"to furfuryl alcohol, 2-( 1,2-O-isopropylidene-1,2-dihydroxyethyl)furan, 1-(2-furyl)ethanoI, and 2-(2-furyl)glycerol 1,3-diacetate. The methyl 2,3-dideoxy-~~-2-enopyranosides can be converted, by way of hydroxylation or epoxidation and subsequent opening of the oxirane ring, into a variety of monosaccharides. If an optically active 2-furylcarbinol is used, it should be possible in some cases to obtain sugars of the D or L series in an optically pure state.122c Making use of the general method outlined in Scheme 50a, Achmatowicz and SzechnerlZzdhave recently synthesized m-cinerulose A (see Scheme 50b); cinerulose A has been isolated from the anthracycline antibiotic cinerubin A, and shown to have the structure of 2,3,6-trideoxy-~-glycero-hexopyranos-4ulose. Treatment of 1-(2-furyl)ethanol with bromine in methanol gave 1- [2-(2,5-dimethoxy-2,5-dihydrofuryl)]et hanol which, on catalytic hydrogenation, afforded I - [2-(2,5-dimethoxy-2,3,4,5-tetrahydrofuryl)]ethanol;the dihydro and tetrahydro derivatives were mixtures of cis and trans isomers. Acid-catalyzed hydrolysis of the tetrahydro derivative gave m-cinerulose A as an anomeric mixture. Alternatively, DL-cinerulose A could be obtained by conversion of the dihydro derivative into 2,3,6-trideoxy-a-~~-hex-2-enopyranos-4-ulose, followed by catalytic hydrogenation. A synthesis of 3-amino-3-deoxy-o~-pentofuranosidederivatives from the 2,5-dihydrofuran derivative 182 has been reported by Iwai et a1.122The key
R' I
0
0
rcduction
+
Additions to double bond or epoxidation and oxirane ring opening
1
Monosaccharides Scheme 5Oa
57
58
The Total Synthesis of Carbohydrates
Scheme Sob
steps were epoxidation of the double bond and opening of the cpoxide ring with ethanolic ammonia. Other syntheses of amino sugars are discussed in detail in Section 7A.
182
7. MISCELLANEOUS SYNTHESES
A.
Amino Sugar Derivatives
Carbohydrates in which a hydroxyl group has been replaced by an amino substituent are called amino sugars (strictly, aminodeoxy sugars). These derivatives are of widespread occurrence in nature. They are found in many polysaccharides and mucopolysaccharides of microbiological and animal origin and in several antibiotics. The majority of amino sugars have been synthesized from readily available monosaccharide derivatives. However, syntheses from noncarbohydrate precursors have been reported. David and V e y r i e r e ~ 'have ~ ~ recently prepared 3-acetamido-2,3-dideoxyD-tetrose (188) from L-aspartic acid by the route shown in Scheme 51. 3-Methoxycarbonyl-3-trifluoroacetamidopropionyl ~hloride'~4(183) was obtained from L-aspartic acid and converted into 3-niethoxycarbonyl-3trifluoroacetamidopropionaldehyde (184) by a Rosenmund reduction. Treatment of 184 with methanolic hydrogen chloride yielded methyl 2amino-4,4-dimethoxybutyrate (185) which, on reduction with lithium
7. Miscellaneous Syntheses
59
0
I
NHCCF,
NHZ
I I
HO~C-CH~-C-CO~H
I I
+ ,ClOC--CH~-C--CO~CH3 H
H L-Aspartic acid
(CH30)2CH-CHz-
ZHa
1
-C02CH3
I
183
t
0 NHLLF,
I I
OHC-CHZ-C-COZCH,
H
H
185
184
NHAc
NHZ
I
(CH3O)ZCH-CHZ-C-CHzOH
I
I I
+ (CHSO)ZCH-CHz-C-CH20H
H 186
TH NHAc 188
Scheme 51
aluminum hydride, afforded 3-amino-4-hydroxybutyraldehyde dimethyl acetal (186); the acetamido acetal 187 was obtained by treatment of a methanolic solution of the amino acetai 186 with acetic anhydride. Acidcatalyzed hydrolysis of 187 yielded the crystalline 3-acetamido-2,3-dideoxy D-tetrose (188). The Akabori has been used by Ichikawa et a1.lZeto prepare 2-amino-2-deoxyaldonic acids. The syntheses involved the base-catalyzed condensation of N-pyruvylideneglycinatoaquocopper(I1) (189) with aldehydes to give the corresponding b-hydroxy amino acids; the reactions proceed at pH 8.0-9.5. Thus reaction of 2,3-O-isopropylidene-uldehydo-~-glyceraldehyde with 189, followed by treatment of the mixture with sodium sulfide, gave a
The Total Synthesis of Carbohydrates
60
crystalline 2-amino-2-deoxy-4,5-O-isopropylidene-~-penton~c acid : COOH
0
\\ C-
I
C
HC=O
N-CH2
HCO
I
1I
0
rc"\ \ J I 0-c
H2O
189
\\
*2H20+
1
\
,CMe,
I I CHOH I
CHNHa pn 93
Na,S
+-
H,CO
1
HCO
\
,CMe*
H2C0
0
A branched-chain amino sugar has been prepared by Kuhn and Weiser.'*' Treatment of the aldehyde 190 (see Scheme 52) with benzylamine and hydrogen cyanide gave the amino nitrile 191. Hydrogenation of 191 over prehydrogenated palladium oxide-on-barium sulfate afforded 2-amino-3,3dimethyl-4-hydroxybutyraldehyde, the furanoid form (192) of which has been established. CEN HC=O CHNHCH2Ph I I + H,C-C-CH, --+ I I H3C-C-CH3 CHB NH2 CH,OH I CH,OH
qoH
190
191
192
Scheme 52
The total synthesis of 4-amino derivatives of ethyl 4-deoxy-a-~L-lyxopyranoside has been recently achieved by Mochalin et al.127n(see Scheme 52a) by way of a Cope reaction. Ethyl 3,4-dideoxy-3-(dimethylamino)-a-~~-lhreopentopyranoside was converted into the corresponding N-oxide by treatment with a 3 % aqueous-methanolic solution of hydrogen peroxide; pyrolysis of the N-oxide gave 2-ethoxy-3,6-dihydr0-2H-pyran-3-01.Oxidation of the olefin with peroxybenzoic acid afforded ethyl 3,4-anhydro-@-~~-ribopyranoside. Treatment of the epoxide with aqueous solutions of ammonia, dimethylamine, and aniline at room temperature gave the 4-amino derivatives of ethyl 4-deoxy-a-~~-lyxopyranoside. B.
Deoxyfluoro Sugar Derivatives
For several years, there has been considerable interest in deoxyhalo sugars not only because of their potential intrinsic value in biochemistry or
7. Miscellaneous Syntheses
61
pharmacology but also because of their utility in the synthesis of other rare sugars such as deoxy and aminodeoxy sugars. The replacement of a hydroxyl group in a monosaccharide by a chlorine or a fluorine atom is particularly interesting to enzymologists, since the size of these halogen atoms is similar to that of a hydroxyl group, but they have very different capacities for forming covalent or van der Waals linkages. Until recently, the only way of producing sugar derivatives containing secondary fluorine was by total synthesis.lZ8The first such derivatives to be prepared were fluoropolyols. Thus Claisen condensation of diethyl oxalate and ethyl fluoroacetate gave diethyl fluorooxalacetate, which, on reduction with lithium aluminum hydridelgQ or sodium borohydride,lgO afforded two racemic fluorinated tetritols, 2-deoxy-2-fluoro-~~-erythritoland 2-deoxy-2fluoro-DL-threitol (Scheme 53; only one isomer of each DL mixture is shown). In a similar way, condensation of methyl 2,3-O-isopropylidene-~~glycerate with ethyl fluoroacetate has led to the preparation of 2-deoxy-2f l u o r o - ~ ~ - r i b i t o By l . ~ standard ~~ procedures, these 2-deoxy-2-fluoroalditols have provided fluoro derivatives of gly~eraldehyde,’~~ glycerol,’32 glyceric acid,1S3 and erythronic
LO -
C0,Et FCH2-COZEt
+
(CO,EOZ
-I
CHF
I
CH,OH
I
I
HCOH HCF
C0,Et Scheme 53
I
CH,OH
CH,OH
I
HCOH
+ I
FCH
I
CH,OH
62
The Total Synthesis of Carbohydrates
C. Branched-Chain Sugars
The first branched-chain sugar was detected in 1901 by Vongerichten in the form of a glycosidic component in parsley. It was given the name apiose, and its structure was later found to be 3-C-hydroxymethyl-~-glycero-tetrose. A synthesis of L-apiose from a noncarbohydrate precursor has already been described (see Section 4, Scheme 19). In 1919 Emil Fischer and Freudenberg detected a branched-chain sugar in hamamelitannin from Hamamelis virginiana which they named hamamelose. I t was later identified as 2-Chydroxymethyl-D-ribose. For a long time, apiose and hamamelose remained the only examples of branched-chain sugars. With the advent of the “antibiotic era,” however, several new branched-chain sugars were discovered. Many of these have been prepared from simple sugar derivatives. In this section syntheses of DL-mycarose and DL-epimycarose from noncarbohydrate precursors are described. Mycarose occurs as a component of a number of macrolide antibiotics. Its structure has been shown to be that of 2,6-dideoxy3-C-methyl-~-ribo-hexose(193). Epimycarose is epimeric with mycarose at c-3.
193
Woodward and his associate~’~5 have synthesized both DL-mycarose and DL-epimycarose (see Scheme 54). Addition of the Grignard reagent derived from propyne to acetoacetaldehyde dimethyl acetal (194) gave the acetylenic alcohol 195, which was converted into the cis olefin 196 by hemihydrogenation over a poisoned palladium catalyst. cis-Hydroxylation of 196 could be effected with osmium tetroxide or with dilute aqueous alkaline potassium permanganate. The mixture of triols 197 and 198 was cyclized in methanolic hydrogen chloride to give a mixture of the racemic glycosides 199 (methyl DL-mycaroside) and 200 (methyl DL-epimycaroside). The relative configurations of the separated glycosides were elucidated by chemical means, and the relationship of racemic 199 to mycarose was demonstrated by its conversion into the 4-0-p-tolysulfonyl derivative of the free sugar, which was spectroscopically indistinguishable from the corresponding natural sugar derivative. Synthetic racemic 199 was resolved by way of its bornanol-10sulfonates. Korte et a1.lSs have prepared DL-mycarose and DL-epimycarose starting
To
+ CH,CrCMgBr
(CH, O ) ~ ~ H
1
194
H3C H 196
197
198
.1
H ,CO
H3C,
\CH
.1
H3C0
199
OH
\
H 200
Scheme 54 63
64
The Total Synthesis of Carbohydrates
from 3-methylhex-3-eno-l
-+
5-lactone (201):
201
The synthesis of DL-mycarose involved reduction of the lactone function with lithium aluminum hydride to the hemiacetal, which was converted into a glycopyranoside. cis-Hydroxylation of the double bond, followed by acid-catalyzed hydrolysis, afforded the free sugar. DL-Epimycarose was obtained by trans-hydroxylation of the appropriate intermediate. Finally, a synthesis of DL-mycarose and DL-epimycarose was achieved by Grisebach et aI.l3' from methyl 3-hydroxy-3-methyl-4-hexenoate (202). Oxidation of 202 with monoperoxyphthalic acid, followed by acid-catalyzed hydrolysis, gave a mixture of the 3-epimeric mycaric lactones (203 and 204),
!
0 - c 7 I I H3C- -C-OH CH2 I
"C-"\
C02CH3
I I H3C-C-OH I HC I CH I
HC-0)
CH2
CH3 202
-+ DL-Mycarose
I
CH3
c
203
0-c
~
I I HO-C-CH, I
CH,
HC-0
I
CH3 204
Scheme 55
--f
m.-Epimycarose
8. Enzymic Syntheses
65
which was reduced by bis(3-methyl-2-butyl)borane to a mixture of DLmycarose and DL-epimycarose (Scheme 55). A separation of the epimeric sugars was achieved by preparative paper chromatography. 8. ENZYMIC SYNTHESES
Hough and Jones19*speculated that, since dihydroxyacetone phosphate will react with D- or L-glyceraldehyde under the influence of the enzyme aldolase to yield D-fructose and L-sorbose 1-phosphates, respectively, then possibly other aldehydes would react in a similar manner. It was shown that glycolaldehyde in the presence of a l d ~ l a s eproduces ~ ~ ~ D-rhreo-pentulose l-phosphate, DL-lactaldehyde yields 6-deoxy-~-arubino- and ~-xylo-hexulose l - p h o ~ p h a t e s , ~and ~ ~ that propionaldehyde gives 5,6-dideoxy-~-fhreohexulose 1-ph0sphate.l~~ In these cases and in almost all other examples known, the two new hydroxyl groups produced have the D-fhreoconfiguration. In these examples the condensation leads in a stereospecific fashion to the formation in each case, of one sugar derivative only from noncarbohydrate material (see Scheme 56). A recent example142of this type of condensation is the observation that pyruvate and formaldehyde, under the influence of an aldolase, yield 3deoxy-Ztetrulosonic acid. Similarly, glyoxylate and pyruvate form a 3deoxy-2-keto-~-pentaric acid (Scheme 57). Uehara and M a t ~ u k a w a 'showed ~~ that D-ribose was produced exclusively CH,OP
L o HoClH
I HCOH I
HOCH
I
CH,OH L-Sorbose I-phosphate
CH,OP
1
CH,OP
I
I
o/do/nse
c=o C=O n/dv/ose I C I H z o H / r H O C H I
HC=O
I HOCH I
CH20H OH
/
p = -p=o
\
OH
Scheme 56
HC=O
HCOH
R
R
I
I
66
The Total Synthesis of Carbohydrates
coo0 L
I
O
CH,
+
coo0 I c=o I
I c=o
I 6 0 I + CH,
CH,
HCHO
coo0
coo0
+
I
CHO
CH,OH
I
I
t-CH,
I I coo0
HCOH
coo0 Scheme 57
when baker’s yeast was added to a solution of D-glyceraldehyde and hydroxypyruvic acid in a phosphate buffer at pH 6.8. Presumably, the a-keto acid, produced by an aldol-type condensation, is decarboxylated to yield the pentose sugar. It is interesting that in this case condensation leads to the production of two new hydroxyl groups with the D-erythro configuration (see Scheme 58).
coo0 I c=o
coo0 -
I c=o I I CH,OH HCOH + + I HC=O HCOH I I HCOH HCOH I I CH,OH
CH,OH-
HC=O
I I + HCOH I HCOH
HCOH
I
CH,OH D-Ribose
Scheme 58
9. SYNTHESIS OF CYCLITOLS
The class of compounds called cyclitols, which includes the inositols (cyclohexanehexols), quercitols (cyclohexanepentols), inososes (pentahydroxycyclohexanones), inosamines (aminodeoxyinositols), and quinic and shikimic acids, has long been of interest to carbohydrate chemists. Many such compounds have been synthesized in nature and in the laboratory from sugar precursors. However, there is a significant literature on their synthesis from noncarbohydrate precursors; some representative examples are briefly discussed in this section.
9. Synthesis of Cyclitols
67
Zelinski et a1.lMobtained the cyclohexanetetrol205 by direct hydroxylation of 1,3-~yclohexadienewith permanganate:
HO
OH 205
A mechanism for this trans-hydroxylation by permanganate has been proposed by Sable.ld6Compound 205 has also been prepared by a Diels-Alder synthesis (see Scheme 59).ld6Reaction of furan with vinylene carbonate gave the adduct 206, which is called 1 ,Canhydro-cis-conduritol carbonate. Successive acidic and basic hydrolysis afforded conduritol-C; or, after
206
,
HO O H
HO OH 205
OH
OH HO OH neo-Inosito1
Condurit 01- C Scheme 59
epi-Inositol
68
The Total Synthesis of Carbohydrates
preliminary hydrogenation, the cyclohexanetetrol 205 was obtained. cisHydroxylation (endo or exo) of the adduct 206, followed by acidic and basic hydrolysis, gave a mixture of rieo- and epi-inositol. The Diels-Alder reaction has also been used for the preparation of alloinositol by Criegee and BecherId7(see Scheme 60). Thus frans,lrans-diacetoxybutadiene and vinylene carbonate condensed to give an adduct 207, which, on hydroxylation by osmium tetroxide, followed by hydrolysis, gave d l o inositol. In this example the osmium tetroxide approaches from the unOAc
I
CH
I1
0
201
ah-Inositol
Scheme 60
hindered side of 207, and the other possible product, cis-inositol, is not formed. Posternak and Friedlil**prepared five cyclohexanetetrols by the appropriate cis- or tram-hydroxylation of cis- or rrans-3-cyclohexene-l,2-diol (see Scheme 61). cis-Hydroxylation may be achieved with permanganate or with silver chlorate-osmium tetroxide,l** whereas trans-hydroxylation occurs with a peroxy acid,l4* or with silver benzoate-iodine (PrCvost reagent).1dg Nakajima and his associates150have also employed various hydroxylation methods in the synthesis of inositols from cis- and trans-3,5-cyclohexadieneI ,2-diols (208 and 209). These starting materials were obtained from a-3,4,5,6tetrachlorocyclohexene by hydroxylation of the double bond, followed by removal of the chlorine atoms with zinc. An early synthesis of inositols from a noncarbohydrate precursor involved hydrogenation of hexahydroxyben~ene.~~~ K u h n et al.lS2repeated this study using a palladium catalyst and tetrahydroxybenzoquinone as starting material. The products of the hydrogenation were fully investigated by Angyal and McHughIS3 using cellulose chromatography. Five inositols, three quercitols, and one inosose (210) were isolated; myo-inositol (211) was the predominant product (17 %). Some cyclohexane-tetrols and -trials were also formed. (-)-Quinic acid, 1,3,4,5-tetrahydroxy-1-cyclohexanecarboxylicacid (212),
c
rHpH HO
HO
HO 'ChOH
P- QHLHPtl
HO
HO
HO
H
QH OH I
.1
HO OH
HoQOH
OH
Scheme 61
I
HO
208
209
HO OH
HO 0H 210
211
cis-Inosose
qw-Inositol
212 69
The Total Synthesis of Carbohydrates
70
has long been known and occurs widely in the plant kingdom. Stereospecific syntheses of ( f)-quinic acid from noncarbohydrate precursors have been achieved. One of these was by Smissman and Oxman,'" and the sequence of reactions is shown in Scheme 62. A Diels-Alder reaction of trans,trans-l,4-
I
//
1I
CH,CO
HC
+
I
HC
\
0
C _ _H-
H
0
II
COCH,Ph
CH
I
CI
213
0
II
0
\ 0
0
I
I
OH
HO 215
216
(f)-Quinic
214
acid Scheme 62
dichlorobuta-l,3-diene with benzyl a-acetoxyacrylate (prepared from benzyl pyruvate by refluxing with acetic anhydride in the presence of p-toluenesulfonic acid) gave the adduct 213, which was converted into the chloro lactone 214 on heating. Compound 214 was cis hydroxylated with osmium tetroxide to give 1-acetyl-6-chloroquinide (215). Hydrogenolysis of the chlorine atom was achieved with W-6 Raney nickel to yield (f)-1-acetylquinide (216), from which (f)-quinic acid could be obtained by heating in aqueous potassium hydroxide solution. Wolinsky et al.155have also synthesized (f)-quinic acid (see Scheme 63). Methyl pyruvate was first converted into methyl a-acetoxyacrylate, which
9. Synthesis of Cyclitols
71
i
Br 218
217
219
(+)-Quinicacid OH 221
220
Scheme 63
was then condensed with 1,3-butadiene to give methyl 1-acetoxy-3-cyclohexene-I-carboxylate (217); hydrolysis of the adduct 217 gave the hydroxy acid 218. Bromo-lactonization of 218 afforded I-hydroxy-5-bromo-3oxabicyclo [3.2.l]octan-2-one (219). Dehydrobromination of 219 was effected by heating it at 130-180" with triethylamine in benzene. Hydroxylation of the unsaturated lactone 220 with osmium tetroxide gave ( f)-quinide (221), which, on hydrolysis, afforded (f)-quinic acid. Smissman et a1.16'J have also achieved the first total chemical synthesis of shikimic acid (222). This compound has been shown to be an important intermediate in aromatic biosynthesis. Irans,trans-l,4-Diacetoxybutadiene was condensed with methyl acrylate to produce cis-3,6-diacetoxycyclohexene4-carboxylate (223) (Scheme 64). The double bond was cis-hydroxylated with osmium tetroxide to give methyl ~-2,~-S-diacetoxy-a-3,a-4-dihydroxycyclohexylcarboxylate (224), which was converted into the 3,4-acetonide. The acetonide was pyrolyzed to give methyl (&)-3-O-acetyI-4,5-O-isopropylideneshikimate (225), which was converted into (f)-shikimic acid by successive acidic and basic hydrolysis.
222
72
The Total Synthesis of Carbohydrates
OAc
OAc 223
225
\
224
DL-Shikimic
acid
Scheme 64
A more recent synthesis of shikiniic acid is by Grewe and Hinrichs's' (Scheme 65). I ,4-Cyclohexadiene-I-carboxylic acid was prepared from 1,3-butadiene and propiolic acid, and converted into the methyl ester 226. Compound 226 was trans-hydroxylated with peroxyformic acid, and the diol was acetylated to give 227. The di-0-acetyl derivative was refluxed in carbon tetrachloride with N-bromosuccinimide, and the product was treated in acetic acid with silver acetate; removal of the acetyl groups finally gave methyl (f)-shikimate (228). Cleophax et a1.157a have obtained methyl (-)-3,4,5-tri-O-benzoylshikirnate by treatment of methyl (-)-3,4,5-tri-0benzoylquinate with sulfuryl chloride-pyridine in anhydrous chloroform. It has recently been shown that cyclitols can be converted into monosaccharides. The key intermediates in this work were seven-membered hemiacetal lactones. The first such synthesis was that of DL-allose from myoino~itol'~*(Scheme 66). DL-1 ,2: 3,4-Di-O-isopropyIidene-5,6-di-O-p-tolylsulfonyl-epi-in~sitol'~~ was obtained from myo-inositol and treated with sodium niethoxide in tetrahydrofuran to give DL-I ,2:3,4-di-O-isopropyIidene-6-0-methyl-epi-inositol (229). Oxidation of 229 with active manganese
9. Synthesis of Cyclitols
73
226
2
228
v
~ C O , C H , AcO 227
Scheme 65
dioxide gave the hemiacetal lactone 230. A more convenient procedure for the preparation of 230 involved oxidation of 229 with the Pfitzner-Moffatt reagent (methyl sulfoxide-N,N'-dicyclohexylcarbodiimide-pyridiniumphosphate) to give the epi-inosose derivative, which was then subjected to the Baeyer-Villiger reaction with peroxybenzoic acid. Compound 230 was heated in methanol containing a catalytic amount of sulfuric acid, and the resultant product was acetylated with acetic anhydride in pyridine; two components were isolated (231 and 232). Reduction of the monoacetate 231 with lithium aluminum hydride gave methyl 2,3-O-isopropylidene-/?-~Lallofuranoside (233). Hydrolysis of 233 afforded DL-allose (234, only the Disomer is shown). DL-Ribose (235) could also be obtained from 233 by successive oxidative cleavage by periodate, reduction with lithium aluminum hydride, and acid-catalyzed hydrolysis. Fukami et al.'"O have used a similar approach to prepare 5 - d e o x y - ~ ~ allose derivatives. In this case the seven-membered hemiacetal lactone 236 was employed. 4-Deoxy-~~-ribose(238) was also obtained from 236 by reduction with lithium aluminum hydride to give 2-deoxy-3,4-O-isopropylidene-DL-allitol (237), followed by oxidative cleavage with periodate and
I
CO2CH3
I
0 .
P
/c\
AcO OAc
H 3C' CH 231
HC=O
w 0.p
I I
HCOH
HOCHO OCH3 t-
HCOH
I
CH,OH 235
Ribose
/ / HSC CH3 233 Scheme 66
74
HC=O
CHZOH
I
I HCOH I HCOH I
\
232
- I
HCOH HCOH
I I
HCOH CH,OH 234
Atlose
References
75
acid-catalyzed hydrolysis:
CH20H
-
I I
HC=O
HCOH
236
LiAIH)
HCO
\
HCO
I
CH2
I
i
I)IO,O
HCOH
2)H@
HCOH
- 1
1 1
CH? CH20H
CH20H 231
238
REFERENCES
1. B. Zuckerrnan, D. Buhl, P. Palmer, and L. Snyder. Aslrophys. J . , 160, 485 (1969). 2. J. Shapka. Cltem. Etg. News, 47 (41), 40 (1969);Sci. Tech. Aerosp. Rep., 6, 2597 (1968). 3. A. Butlerow, Contpf.r e d , 53, 145 (1861);A. Butlerow, Ann., 120, 295 (1861). 4. C.H. Riesz, Sci. Tech. Rep, 6,957 (1968); Client. A h . , 96815 (1969). 5. E. Fischer, Ber., 21,989 (1888);23, 389,2144 (1890). 6. 0.Low, J. Prakf. Cheni., [2]33, 321 (1886). 7. 0.Low, Ber., 22, 471 (1889). 8. E. Schmitz, Ber., 46, 2327 (1913); W. Kuster and F. Schoder., Hoppe-Seyler'sZ. Physiol. Chern., 141, 110 (1924). 9. E. Fischer and J. Tafel, Ber., 20, 1088,2566 (1887). 10. E. Fischer and J. Tafel, Ber., 20, 3384 (1887). 11. H. 0.L. Fischer and E. Baer, Heh. Chitn. Actu, 19, 519 (1935); 20, 1213 (1937). 12. B. L.Horecker,J. Cell. Conip. Physiol., 54 (Suppl. l), 89 (1959);E.Grazi, T. Cheng, and B. L. Horecker, Biocheni. Biophys. Cottinrun., 7 , 250 (1962). 13. W. G. Berl and C. E. Feazel, J. Am. Chcni. Soc., 73, 2054 (1951);C.D.Gutsche, R. S. Buriks, K. Nowotony, and H. Grassner, J. Ant. Cheni. Soc., 84,3775 (1962). 14. R. Breslow. Terruhedroti Left., 1959, 22. 1960, 886. 15. A. A. Marei and R. A. Raphael, J. Chem SOC., 16. E.Pfeil and H. Ruckerl, ,4nji.,641, 121 (1961);J. K.N.Jones,personal observations. 17. H. Ruckert, E. Pfeil, and G. Scharf, Ber.. 98,2558 (1965). 18. L. Hough and J. K. N. Jones, J. Client. Soc., 1951, 1126. 19. E. Mariani and G. Torraca, h i t . Siigur J., 55, 309 (1953); Chetn. A h . , 48, 4869 (1954).
76
The Total Synthesis of Carbohydrates
K. D. Partridge, A. H. Weiss, and D. Todd, Abstracts, 163rd National Meeting of the American Chemical Society, Boston, Mass., April 1972,CARB 2. See also A. H. Weiss and J. Shapira. Hydrocarbon Process., 49, 119 (1970);T.Mizuno. T. Mori, N. Shiomi, and H. Nakatsuji, J. Ax'. C h t i . SOC. Jup.. 44, 324 (1970);T.Mizuno, M. Asai, A. Misaki, and Y . Fujihara, ibid., 45, 344 (1971). 19b. H.Tarnbawala and A. H. Weiss, Abstracts, 163rd National Meeting of the American Chemical Society, Boston, Mass., April 1972. PETR 8. 20. J. K.N. Jones, unpublished results. 21. C. Neuberg, Ber., 35,2626 (1902). 22. C.D. Hurd and J. L. Abernethy, J . A M . Clieni. Soc., 63, 1966 (1941). 23. L. Hough and J. K. N. Jones, Nature, 167, 180 (1951);J. Cherri. SOC.,1951, 1122, 3191. 23a. L. Tidwell, J. Lecocq, H . B. Chermside, and J. Shapira, Proc. WesIer/r Phor/iiacol. Soc., 13, 30 (1970). 24. L. M.Utkin, Dokl. Akod. Nuirk. S.S.S.R., 67, 301 (1959). 25. A. Dmytraczenko, J. K. N. Jones, and W. A. Szarek, unpublished results. 26. F. Shafizadeh, Ado. Corbohyd. Chetti., 11, 263 (1956). 27. L. M.Utkin, Zh. Obshchei Khirti., 25, 530 (1955);Cliem. U.S.S.R.,25, 499 (1955). 28. C.Griner, A ~ J IC/iitii. I. Phys. [6],26, 369 (1892). 29. R. Lespieau and J. Wiemann, Bull. SOC.C/iet/i. [4J, 53, 1107 (1933);J . Wiertiarrn, Anti. Chim (111, 5 , 267 (1936). 30. Cf. R. Lespieau, Adu. Curbohyd. Chein., 2 , 108 (1946). 31. R. Lespieau and J. Wiemann, C o t t y f .rend., 194, 1946 (1932). 32. R. Lespieau, Bull. SOC.Chitti. (41,43, 204 (1928). 33. R. Lespieau, Eiill. SOC.Chirti. [4].43, 657 (1928). 34. R. A. Raphael, J. Cheni. SOC.,1949,S44. 35. 1. lwai and T . Iwashige, Client. Pliarni. Bitll. (Tokyo), 9, 316 (1961); T. Iwashige, C h e w Phanii. B~ill.(Tokyo),9, 492 (1961). 36. 1. lwai and K. Tomita, Chem. Pharni. Birll. (Tokyo),9, 976 (1961). 37. 1. Iwai and K. Toniita, Cheni. Pliarrii. Eidl. (Tokyo), 11, 184 (1963). 38. 1. Iwai, T. Iwashige, M. Asai, K. Tomita, T. Hiraoka, and J. lde, C/ieni. P/inrm. Eir/l. (Tokyo), 11, 188 (1963). 39. L. Hough, Chem. hid. (London), 1951,406. 40. W. G.Overend and M . Stacey, J. Sci. Food Agr., 1, 168 (1950). 41. M. M. Fraser and R. A. Raphael, J. Cheni. SOC.,1955,4280. 42. F.Weygand and H. Leube, Eer., 89, 1914 (1956). 43. G.Nakaminami, M.Nakagawa, S . Shioi, Y.Sugiyama, S. Isemura, and M. Shibuya, TefralredroriL e f t . ,1967, 3983. 44. C. C. Price and R. B. Balsley, J. Org. Chetti., 31, 3406 (1966). 45. R. A. Raphael, J. Chem. Soc., 1952,401. 45a. W.F. Beech, J. Clieni. SOC.,1951, 2483; B. W. Horrom and H. E. Zaugg, J. Ant. 79, 1754 (1957). Cherri. SOC., 45b. J. Kiss and F. Sirokrnhn, Helv. Chitti. ACIU,43, 334 (1960). 46. D.J. Walton. Can. J. Cherti., 45, 2921 (1967).
19a.
References 47. 48. 49. 50. 51. 52.
53. 54.
55. 56. 57. 58. 59.
60.
61. 62. 63. 64. 65.
66.
67. 68.
69. 70. 71. 72.
73. 74. 75.
77
D. Horton, J. B. Hughes, and J. K . Thomson, J. Org. Chem., 33,728 (1968). E. Fischer, Bey., 22,2204 (1889). H. J. Lucas and W. Baumgarten, J. Am. Chenr. Soc., 63, 1653 (1941). H. J. Bestmann and R. Schmiechen, Eer., 94,751 (1961). F. Weygand and R. Schmiechen, Ber., 92, 535 (1959). A. J. Ultke and J. B. J. Soons, Rec. Trau. Chim. Pays-Bas, 71, 565 (1952). R. L u k e , J. Jary, and J. Nernec, COILCzech. Chem. Commun., 27, 735 (1962). S. Sasaki, J. Chem. SOC.Japan, Pure Chem. Sect., 78, 1464 (1957); Cliem. Zentr., 1958,9467. J. Jary and K. Kefurt, Coll. Czech. Chem. Commun., 27, 2561 (1962). K. Koga, M. Taniguchi, and S.Yamada, Tetrahedron Left., 1971,263. B. Belleau and Y.-K. Au-Young, J. Am. Chem. SOC.,85,64 (1963). For reviews see H. Paulsen, Angeiv. Chem. fnt. Ed., 5 , 495 (1966); H. Paulsen and K. Todt, Adv. Carbohyd. Chem., 23, 115 (1968). D. M. Vyas and G. W. Hay, Chem. Commun., 1971, 1411; Can. J. Chem., 49, 3755 (1971). For a recent review on acetals and hemiacetals see: E. Schmitz and I. Eichhorn, in S. Patai, Ed., The Chemistry offhe Ether Linkage, Interscience, New York, 1967, Chapter 7. S. J. Angyal, Atrgeiu. Chem. lot. Ed., 8 , 157 (1969). C. L. Wilson, J. Chem. SOC.,1945, 52. H. Normant, Compt. rend., 228, 102 (1949). P. Dimroth and H. Pasedach, Angeiv. Chem., 72,865 (1960). R. Paul, M. Fluchaire, and G. Collardeau, Bull. SOC.Chim. A.,1950,668; M. L. A. Fluchaire and G. Collardeau, U.S. Patent 2,556,325 (1951); Chem. Abs., 46, 1046 (1952). L. E. Schniepp and H. H. Geller, J. Am. Chem. SOC.,68, 1646 (1946). W. J. Gensler and G. L. McLeod, J. Org. Cheni., 28,3194 (1963). K. Alder, H. Oppermans, and E. Ruder, Ber., 74, 905,920,926 (1941). C. W. Smith, D. G. Norton, and S. A. Ballard, J. Am. Chern. SOC.,73,5267,5270, 5273 (1951). E. Dyer, C. P. J. Glaudemans, M. J. Koch, and R. H. Marchessault, J. Chenr. SOC., 1962, 3361. C. B. Anderson and D. T. Sepp, Tetrahedron, 24, 1707 (1968); D. T. Sepp and C. B. Anderson, Tetrahedron, 24, 6873 (1968). R. U. Lemieux and N. J. Chu, Abstracts, 133rd National Meeting of the American Chemical Society, San Francisco, Calif., April 1958, p. 31N; E. L. Eliel, N. L. Allinger. S. J. Angyal, and G. A. Morrison, ConformafionalAnulysis, Interscience, New York, 1965, p. 375. S. Wolfe, A. Rauk, L. M. Tel, and I. G.Csizmadia, J. Cheni. Soc., (B), 1971, 136. U. E. Diner and R. K.Brown, Can. J. Chem., 45,2547 (1967).
E. L. Eliel, B. E. Nowak, R. A. Daignault, and V. G. Badding, J. Org. Chem., 30, 2441 (1965).
78
The Total Synthesis of Carbohydrates
76. S. 0. Lawesson and C. Berglund, Ark. Kemi, 17, 475 (1961); G. Sosnovsky, J. Org. Chem., 25, 874 (1960); Teirahedrorr, 13, 241 (1961). 77. C. D. Hurd and C. D. Kelso, J. Am. Chem. Soc., 70, 1484 (1948). 78. F. Sweet and R . K. Brown, Can. J. Chem., 45, 1007 (1967). 79. F. Sweet and R. K. Brown, Can. J. Chem., 44, 1571 (1966). 80. R. Paul, Bull. SOC.C/iim. Fr., 1 [V], 1403 (1934). 81. R. Paul, Conpr. rend., 218, 122 (1944). Chin,. Fr., 1951, 829. 82. 0. Riobt, Bull. SOC. 83. R. U. Lernieux and B. Fraser-Reid, Can. J. Chetn., 43, 1460 (1965). 84. W. Reppe and co-workers, Ann., 596, 86 (1955). 85. A. Senning and S.-0. Lawesson, Teirahedron, 19, 695 (1963). 86. M. J. Baldwin and R. K. Brown, Can. J . Chenr., 45, 1195 (1967). 87. M. J. Baldwin and R. K. Brown, Can. J. Chem., 46, 1093 (1968). 88. F. Sweet and R. K. Brown, Can. J. Cliem., 46, 707 (1968). 89. F. Sweet and R. K. Brown, Can. J . Cheni., 46, 1592 (1968). See also A. Banaszek and A. Zamojski, Roczniki Chem., 45, 391 (1971). 90. F. Sweet and R . K.Brown, Cari. J . Chem., 46, 1543 (1968). 91. M. J. Baldwin and R. K. Brown, Can. J . Chew, 47, 3099 (1969). 92. M. J. Baldwin and R. K. Brown, Can. J. Chem., 47, 3553 (1969). ~ ,(1950); R. M. Srivastava, 92a. G . F. Woods and S . C. Ternin, J . Am. Chem. S O C . , ~139 F. Sweet, 1’. P. Murray, and R. K. Brown, J . Org. Cheni., 36, 3633 (1971). 92b. R . M.Srivastava, F. Sweet, and R. K. Brown, J . Org. Chent., 37, 190 (1972). 93. M. Cahu and G. Descotes, Bull. Soc. Chitn. Fr., 1968, 2975. 94. A. Konowal, J. Jurczak, and A. Zamojski, Roczniki Clieni., 42, 2045 (1968). 95. A. Zamojski, A. Konowal, and J. Jurczak, Roczniki Chenr., 44, 1981 (1970). 96. J. Jurczak and A. Zamojski, Roczniki Chem., 44, 2257 (1970). 97. J. Jurczak, A. Konowal, and A. Zarnojski, Roczniki Chem., 44, 1587 (1970). 98. A. Konowal, A. Zamojski, M. Masojidkova, and J. Kohoutova, Roczniki Chem., 44, 1741 (1970). See also A. Banaszek and A. Zamojski, ibid., 45, 2089 (1971). 99. A. Zamojski, M. Chmielewski, and A. Konowal, Tetrahedron, 26, 183 (1970). 99a. M.Chmielewski and A. Zamojski, Rocrniki Chem., 45, 1689 (1971). 99b. A. Konowal and A. Zamojski, Roczniki Cheni., 45, 859 (1971). 100. Z . Zwierzchowska and A. Zamojski, Roczniki Client., 44, 1609 (1970). 101. R. J. Ferrier, N. Prasad, and G. H.Sankey, J. Chen?.SOC.( C ) , 1968, 974. 102. F. Sweet and R . K. Brown, Cari. J. Client., 46, 2283 (1968). 103. R. M. Srivastava’and R. K. Brown, Cari. J. Chem., 48, 2334 (1970). 104. R. M. Srivastava and R. K. Brown, Can. J . Chem., 48, 2341 (1970). 105. F. Sweet and R. K. Brown, Can. J . Chem., 46, 2289 (1968). 106. T. P. Murray, C. S. Williams, and R. K. Brown, J. Org. Cliem., 36, 1311 (1971). 107. U. P. Singh and R. K. Brown, Can. J . Chem., 48, 1791 (1970). 107a. U. P. Singh and R. K. Brown, Can. J. Chenr., 49,3342 (1971). 108. U. P. Singh and R. K. Brown, Can. J . Chenl., 49, 1179 (1971).
References 109. 110. 111. 112. 113.
T.P. Murray, U. P. Singh, and R. K. Brown, Can. J. Chenr., 49,2132 (1971). R. M. Srivastava and R. K.Brown, Con. J. Clrem., 49, 1339 (1971). F. Korte, A. Bilow, and R. Heinz, Tetrahedron, 18, 657 (1962).
79
H. Newman, J . Org. Chemr., 29, 1461 (1964). V. B. Mochalin, Yu. N. Porschnev, and G. I. Samokhvalov, Zh. Obshch. K h r . , 39, 701 (1969); Chem. A h . , 71, 39346h (1969). 114. S. Yasuda and T. Matsumoto, Tetrahedron Left., 1969,4397. 115. S . Yasuda and T. Matsumoto, Tetrahedron Lett., 1969,4393. 116. Y . Suhara, F.Sasaki, K. Maeda, H. Urnezawa, and M. Ohno, J. Am. Chem. Soc., 90,6559 (1968). 117. S.Yasuda, T.Ogasawara, S. Kawabata, I. Iwataki, and T. Matsumoto, Tetrahedron Lett., 1969, 3969. 118. M. Nakajima, H.Shibata, K.Kitahara, S.Takahashi, and A. Hasegawa, Tetrahedron Left., 1968, 2271. 119. Y. Suhara, K.Maeda, and H. Umezawa, J . Antibiotics, M A , 187 (1965). 120. R. M. Srivastava and R. K.Brown, Can. J. Clrenr., 48, 830 (1970). 121. R. Lukes, M. Moll, A. Zobacova, and J. Jary, Coll. Czech. Chem. Coiirwrin., 27, 500 (1962). 122. I. Iwai, T.Iwashige, and M. Asai, Clrem. Abs., 65, 3950 (1966). 122a. 0.Achmatowicz, Jr., P. Bukowski, B. Szechner, Z. Zwierzchowska, and A. Zamojski, Tetrahedron, 27, 1973 (1971). 122b. 0.Achmatowicz, Jr., and B. Szechner. Tetrahedron Lett., 1972, 1205. 122c. 0.Achmatowicz, Jr., and P. Bukowski, Bull. Acad. Pol. Sci., Ser. Sci. chi ti^., 19, 305 (1971). 122d. 0.Achrnatowicz, Jr., and B. Szechner, Bull. Acad. Pol. Sci.. Ser. Sci. Clrinr., 19, 309 (1971). 123. S. David and A. VeyriJxes, Carbohyd. Res., 13, 203 (1970). 124. Y. Liwschitz, R. D. Irsay, and A. 1. Vincze, J. Chevr. SOC.,1959, 1308. 125. M. Sato, K. Okawa, and S. Akabori, Bull. Cheni. SOC.Jap., 30, 937 (1957). 126. T. Ichikawa, T.Okamoto, S. Maeda, S. Ohdan, Y.Araki, and Y. Ishido, Tetrahedron Lett., 1971,79. 127. R. Kuhn and D. Weiser, Ann., 602, 208 (1957). 127a. V. B. Mochalin, Z. I. Smolina, and B. V. Unkovskii, Zlr. Obshch. Klrirn., 41, 1863 (1971). For a recent review o n fluorocarbohydrates see P. W. Kent, Chem. Ind. (London), 128. 1969, 1128. 129. N. F.Taylor and P. W. Kent, J. Clrem. SOC.,1956,2150. 130. J. E.G.Barnett and P. W. Kent, J. Clrem. Soc., 1963,2743. 131. P. W.Kent and J. E. G. Barnett, J. Clrem. SOC.,1964,2497. 132. N.F. Taylor and P. W . Kent, J. Chem. SOC.,1958,872. 133. A. Bekoe and H. M. Powell, Proc. Roy. Soc., 250A, 301 (1959). 134. R. C. Cherry and P. W . Kent, J. Cltem. Soc., 1962,2507. 135. D. M. Lemal, P. D. Pacht, and R. B. Woodward. Terrulredrorr, 18, 1275 (1962).
80
The Total Synthesis of Carbohydrates
136. 137. 138. 139. 140. 141.
F. Korte, U. Claussen, and K.Gohring, Tetruhedrori, 18, 1257 (1962). H. Grisebach, W. Hofheinz, and N. Doerr, Ber., 96, 1823 (1963). L. Hough and J. K. N. Jones, Ado. Curbolryd. Cherii., 11, 185 (1956). L. Hough and J. K. N. Jones. J . Chenr. Soc., 1952,4047. L. Hough and J. K. N. Jones, 1.Chem. SOC.,1952, 4052. T. C. Tung, K. H. Ling, W. L. Byrne, and H. A. Lardy, Eiochinr. Bioplys. Acra,
14, 488 (1954). 142. R. S. Lane, A. Shapley, and E. E. Dekker, Eioclrein., 10, 1353 (1971). 143. K. Uehara and T. Matsukawa, Jap. Patent 2279 (1959); Cheni. Abs., 54, 13016 (1960). 144. N. D. Zelinski, J. 1. Denisenko, and M. S. Eventova, Conpt. rend. Acud. Sci. URSS. 1, 313 (1935); C/mi. Zeritr., 106, 11, 3765 (1935). 145. H. Z. Sable, Abstracts, 149th National Meeting of the American Chemical Society. Detroit, Mich., April 1965, p. 19C. 146. Y. K. Yurev and N. S. Zefirov, Zli. Obsch. Khim., 31, 685 (1961); Cheiir. Abs., 55. 24573 (1961). Seealso N. S. Zefirov, Y.Yurev, L. Prikazchilova, and M.Bykhovskay Zh. Obsch. Khittr., 33, 2153 (1963). 147. R. Criegee and P. Becher, Ber., 90,2516 (1957). 148. T. Posternak and H. Friedli, Helu. Chiiii. Aclu, 36, 251 (1953). 149. G. E. McCasland and E. C. Horswill, J. Arii. Clievr. Soc., 76, 1654 (1954). 150. S. Takai, M . Nakajima, and I. Tomida, Eer., 89, 263 (1956); M. Nakajima, I. Tomida, and S. Takei, Bet., 90,246 (1957); 92,163 (1959); M. Nakajima, 1. Tomida, N. Kurihara, and S . Takei, Eer., 92, 173 (1959). 151. H. Wieland and R. S. Wishart, Ber., 47, 2082 (1914). 152. R. Kuhn, G. Quadbeck, and E. Rohm. A m . , 565, 1 (1949). 153. S. J. Angyal and D. J. McHugh, J . Cliein. Soc., 1957, 3682. 154. E. E. Smissman and M. Oxman, J. Arir. Cheni. Soc., 86, 2184 (IY63). 155. J. Wolinsky, R. Novak, and R. Vasileff, J . Org. Clrern., 29, 3596 (1964). 156. E. E. Smissrnan, J. T. Suh, M. Oxman, and R. Daniels, J . Atii. Clretn. SOC.,81,2909 (1959). 157. R. Grewe and 1. Hinrichs, Eer., 97,443 (1964). 157a. J. Cleophax, D. Mercier. and S. D. Gkro, Airgew. C/retii.h i t . Ed., 10, 652 (1971). 158. H. Fukami, H.-S.Koh, T. Sakata, and M. Nakajima, Tetrulredrori Lett., 1967,4771. 159. S. J. Angyal and P. T. Gilharn, J . C/rettr. Soc., 1957, 3691. 160. H. Fukami, H.S. Koh, T. Sakata, and M. Nakajirna, Tetra/redrori Lett., 1968, 1701.
Total Synthesis OfNatural Products, Volume1 Edited by John Apsimon Copyright © 1973 by John Wiley & Sons, Inc.
The Total Synthesis of Prostaglandins U. AXEN, J. E. PIKE,
AND
W. P. SCHNEIDER
The Upjohn Company, Kalamazoo, Michigan 1. Introduction
2. Synthetic Routes to Prostaglandins E and F
A. General Strategy B. Stereochemical Principles C. Individual Syntheses 3. Partial Syntheses of PGE, and PGEza from 15(R)-PGA2 4. Synthetic Routes to Structurally Simplified Prostanoic Acids A. d,l-13,14-Dihydro-PGE1 B. 15-Dehydro-PGEl C. 11-Desoxy Prostaglandins D. Cyclopentenone Prostanoic Acids E. d,l-PGE, Methoxime 5. Oxaprostaglandins References
81 89 89 91
96 110
118 118
122 124 125 132 133 138
1. INTRODUCTION
The prostaglandins were discovered in 1933 to 1934 by Goldblatt in England and von Euler in Sweden. Smooth muscle stimulating and vasodepressor activities were found in extracts of human seminal fluid and sheep vesicular glands. von Euler showed that the activity associated with the lipid soluble fractions was not identical with any of the humoral agents known at that 81
82
The Total Synthesis of Prostaglandins
time and named the material prostaglandin. The next major development in the field was the isolation many years later of pure prostaglandin E, and Flx by Bergstrom and his associates and later their classic work on the structure elucidation of these acidic l i p i d ~ . ~They ~ # ~are * carboxylic acids with a 20carbon structure incorporating a 5-membered ring and the hypothetical parent structure is designated prostanoic acid (Fig. 1).
Figure 1. Prostanoic acid
The structures of naturally occurring materials are readily divided into four basic families: PGE, PGF, PGA, and PGB (Fig. 2). The various families include different structures, for example, PGF,a, PGF2a, and PGF,o( with, respectively, 13-trans, 5-cis-1 3-trans, and 5-cis-1 3-trans-17-cis double bonds. All six members of the prostaglandins of the E and F series are known as primary prostaglandins. The structures of the prostaglandins were established by Bergstrom and associates in Sweden by methods that included classical degradation and X-ray crystallographic studies on suitable derivatives.’e2 A noteworthy feature of the structure elucidation was the extensive use of gas chromatography in combination with mass spectroscopy to identify the molecular fragments formed in various chemical degradations. The absolute configuration of PGE, has also been determined.BsThe presence of 19-hydroxy prostaglandins of the A and B classes was shown by Hamberg and Samuelsson in human seminal plasma.63Various earlier reviews cover the structural elucidation and associated chemistry.13~20~B5~s9 The chemical synthesis of the prostaglandins offers an unusually significant challenge to organic chemists for several reasons. It is now well established that the prostaglandins do not occur in male tissues only, but are found in low concentrations in nearly all organs.29 The biological potency and diversity of the prostaglandins is remarkable and several areas are under intensive study to elucidate a possible physiological role for these m a t e r i a l ~ . ~ * ~ ~ 7 Additionally, the possible clinical significance of prostaglandins is underThe congoing serious evaluation, particularly in the control of fertilit~.~? centration of the natural prostaglandins in most tissues is less than 1 pg/g; one exception is in human seminal fluid where concentrations reach about 50-60 pg/ml. Clearly, then, extraction of various organs will not provide useful quantities of the agents. Until very recently prostaglandins had not been isolated from nonmammalian sources in amounts that would serve as potential precursors of the natural hormones. A report by Weinheimer and Spraggins that prostaglandins could be found in a horny coral or gorgonian
N
4
84
The Total Synthesis of Prostaglandins
called Plexaura honrornalla11a-120 has suggested a possible alternative source. Nevertheless, chemical total synthesis with its inherent flexibility seems to offer the best route to obtain these materials and possible structural analogs, which may be needed as therapeutic agents. Before discussing the chemistry of the prostaglandins a brief review will be given of the biosynthesis and metabolism of prostaglandins. It was shown in 1964 that the essential fatty acids were enzymatic precursors of prostaglandin~.~~*~ Bergstrom ~ , ~ 9 ~ 5 0and co-workers in Stockholm and van Dorp and co-workers in Holland demonstrated that 8,1 I , 14-eicosatrienoic acid was the biological precursor of PGE,, arachidonic acid was the precursor of PGE2, and 5,8,11,14,17-eicosapentaenoic acid led to PGE, (Fig. 3). In enzymes derived from guinea pig lung it was shown that arachidonic acid led to PGF,E.~ The prostaglandin-synthesizing enzymes appear widely distributed in various animal tissues.101-102 Studies on the mechanism of the biosynthesis, especially by Samuelsson, have established the mechanism indicated in Fig. 4 involving an intermediate endoperoxide.’OZ The isolation of intermediates such as this have not yet been reported and would represent extremely intriguing synthetic objectives for an organic chemist. Experiments also have been described by workers at Unilever on the nonenzymatic conversion of 8,l I ,14-eicosatrienoic acid to PGE,.as The present very low overall yield in this route suggests this is not a preferred way to obtain prostanoic acids. Chemical synthesis of many of the proposed intermediates in the biosynthesis, especially with specific C14and tritium labeling, may be necessary in the future to delineate exactly the sequence of steps in the biosynthetic formation of prostaglandins. Workers at the Unilever Laboratories have studied the substrate specificity of the enzyme system and these investigations have led to the synthesis of prostaglandins with varied side chain lengths and varied double bond i s o m e r ~ . ~ ’A- ~particularly ~4 important finding was the correlation between the essential fatty acid activity of the precursor acids and the rate of formation of the biologically active prostaglandins. This work has strongly supported the idea that part of the essential function of the unsaturated fatty acids must be associated with their conversion to prostaglandins. Another area of research which has attracted attention involving the biosynthesis has been the report of specific inhibitors of the synthetase. Particularly significant was the activity reported by Nugteren and co-workers for the 8-cis-l2-rrans-lCcis-eicosatrienoicacide7and by Downing for 5,8,11, 14-eicosatetraynoic acid.3*4rszRecent studies by Sih have concentrated on understanding the control mechanisms in the prostaglandin biosynthesis and factors regulating their formation.”’ Particularly intriguing is why the endoperoxide under differing physiological stimuli can produce either the PGE, PGF, or I I-dehydro-PGF structures.’l2 Recently Pace-Asciak and
.
C
O
O
H
‘0H
OH
COOH
w
PGE,
8,11,14-Eicosatrienoic acid
PG Fi U
OH T
C
--
O
O
‘0H PGE2
H
5,8,11,14-Eicosatetraenoic acid
/
OH
“OH PGF2u -
--
COOH
--5,8,11,14,17-Eicosapentaenoic acid
/
\
OH
+OH
OH Figure 3
‘OH PGEo
.-
“OH PGF~u
85
It X 0
I 0
'I
X 0
IIIIIX
I11110
1. Introduction
87
Wolfe described some additional prostanoic acid by-products formed from arachidonic acid.soJJ' Studies, particularly by Sameulsson, h g g b r d , and co-workers, have delineated the principal metabolic inactivation processes both in vitro and in vivo for the natural prostaglandins.8 Especially important is the 15hydroxydehydrogenase enzyme originally isolated from lung t i s s ~ e . ~ -This '~ enzyme, which has been purified, has a high degree of structural specificity for prostaglandin^.^^*^^ The resulting 15-keto compounds appear to be much less biologically active than their 1S(S)-hydroxy p r e ~ ~ r ~ 0 rOther . 1 ~ * ~ ~ ~ ~ metabolic transformations include the reduction of the 13,14-double bond of the A13-15-keto metabolites, the p-hydroxylation and cleavage of the carboxy side chain, and the w- and w-1-hydroxylation of the alkyl (CI3-C2,) side chain.101*102The major urinary metabolites in man of both PGE, and PGF,cr have been identified (Fig. 5).61*65 More recently, Hamberg and Israelsson have described liver enzymes in vitro which convert the 9-keto prostaglandins both to the 9a- and 9p-hydroxy is0mers.6~This conversion of the PGE to the PGFa compounds in vitro is the first reported interconversion of these two series and the significance of this transformation in vivo remains to be evaluated. Enzymatic transformations have also been described which convert the saturated 15-keto compounds to the corresponding 15-hydroxy (S and R) metabolites.64 It also now appears that the 8-is0 compounds are formed in enzymatic transformation^.^^ A particularly important aspect in the further understanding of prostaglandin metabolism is the development of very sensitive assays for prostaglandins. A recent publication by Levine has suggested that the radioimmunoassay technique may be applied to the prostaglandins.'' Samuelsson and Sweeley have developed a new reverse isotope technique which promises to allow the GLC mass spectroscopic measurement of physiological levels of prostaglandin~.~~~ Before discussing in detail the various synthetic approaches, some discussion is in order about the known chemistry of the prostaglandins, particularly the stereochemical features of the molecules. No really systematic studies have appeared detailing the basic chemistry of the molecule, although earlier work has covered general aspects of prostaglandin physicochemical properties. The stereochemistry of substituents has been designated a or p by analogy with steroid nomenclature; a-substituents are oriented on the same side of the five-membered ring as the carboxy (C, to C,) side chain and psubstituents are above the plane of the cyclopentane ring and on the same side as the CIa to C,, side chain. The hydroxyl group at C,, has the 15(S) stereochemistry in the natural mammalian prostaglandins. In earlier publications this was also designated at 15a. The epimer of this hydroxyl group 15(R) has also been named 15s or 15-epi. The stereochemistry of the attached
2 3: 0
0
88
‘T 0
2. Synthetic Routes to Prostaglandis E and F
89
side chain at C8 and Clz was established principally by the earlier X-ray studies of Abrahamsson and co-workers.'Oa The corresponding 8-iso compounds, that is, those with cis-oriented side chains, have also been described:? but the equilibrium occurs largely in favor of the natural configuration and under mild basic conditions the ratio is about 9: 1. When the PGE or PGA compounds are treated with base, the PGB compounds or doubly unsaturated ketones are formed which have a characteristic UV absorption at 278 nm.25 Under mild acidic treatment the PGE compounds can be converted to the PGA c o r n p o ~ n d s . ~Conditions ~ * ~ ~ - ~ ~which have been described include acetic acid and hydrochloric acid buffered with tetrahydrofuran. When prostaglandins are treated with stronger acids, for example, formic acid, epimerization of the C,,-hydroxyl occurs giving a mixture of both 15(R)and 15(S) derivative^.^^ Reductions of the ketone groups at CQ and C,, have been described, and in both cases mixtures of hydroxyl epimers were obtained. ~ 4 . ~ 4 Recently methods have been described for converting the PGA series back to the PGE and PGF prostaglandins via intermediate epoxy ketones (see below).31 The trichloroethyl ester of PGEl could be reconverted to the parent acid by treatment with zinc and acid under conditions which did not affect the p-hydroxy ketone system of the PGE compo~nds.~4 The 9-ketone of PGEl has been converted to an oxime from which the parent ketone could be regenerated using nitrous a~id.4*.~4 The reduction of the carboxyl group to a primary alcohol has been described for both the PGE48*94 and PGFEooo3 series.
2. SYNTHETIC ROUTES TO PROSTAGLANDINS E AND F
A. General Strategy
Total synthesis of the prostaglandin molecule constitutes a formidable challenge to the organic chemist, not as much because of the nature of the different functional groups as because of the numerous asymmetric centers and the relative position of the functional groups to each other. .PGEl (Fig. 2) has four asymmetric centers, an allylic alcohol with trans geometry of the double bond, and a 8-ketol system which is sensitive to acid and base. Acid treatment causes elimination of water to form the a,&unsaturated enone system, PGAl, and base treatment causes elimination and' double bond migration to the fully conjugated PGB s y ~ t e m(Fig. ~ ~ 6). . ~ Therefore ~ the /I-ketol system is generated at a late stage in all the syntheses of E-prostaglandins published
90
The Total Synthesis of Prostaglandins
PG n
PG A
PGE
Figure 6
so far. Both possibilities have been realized: introduction of the 1 1-hydroxyl group at the end of the synthesis or generation of the 9-keto group as the last step. PGF,a (Fig. 2) is more stable than PGEl because the p-ketol unit is reduced to a 1,3-diol, but a new asymmetric center is introduced at C-9. PGE, and PGFza (Fig. 2) display an additional double bond at c5.6; PGE, and PGF3a (Fig. 2) display two additional double bonds at c 5 , 6 and Cl,,18; all have cis geometry. All three E-prostaglandins have been converted to the corresponding PGF compounds by selective reduction of the keto function, but in these cases the corresponding 9p-isomers are formed in addition24.25~62r (Fig. 7).
PG E
PG Fa
PG Fp
Figure 7
Because of these known conversions any synthesis of a PGE constitutes at the same time a synthesis of the corresponding PGF, PGA, and PGB compounds. A recent synthesis3’ by Corey designed to give PGF,a directly avoids the formation of the undesired ,&isomer; because of the use of suitable protecting groups at C-11 and C-15 this approach can still be used for the synthesis of PGE,. In most other cases PGF’s are obtained by reduction of the corresponding PGE’s. Most syntheses start with a suitably functionalized five-membered ring like cyclopentadiene, norbornadiene, or an indanol. Later formation of the ring by cyclization has been utilized, too, but no approach has yet been reported which generates the /?-ketol unit and the ring simultaneously by aldol condensation of the corresponding keto-aldehyde.
2. Synthetic Routes to Prostaglandis E and F
91
B. Stereochemical Principles Stereochemical control at C-8 and C-12 generally does not cause too many problems since the more stable trans configuration of the two side chains is the natural configuration. In most cases where the side chains are introduced subsequently the trans configuration is formed preferentially. Another possibility in the case of PGE's is the isomerization at C-8, which proceeds under basic conditions mild enough not to cause formation of PGA's or PGB's (Fig. 8).47 More difficult is stereochemical control at C-9 of the PGF's.
k= L
0
OH
OH
OH
bH
PGE
8-iso-PGE
Figure 8
As mentioned, reduction of the keto-function produces mixtures of the desired a-isomer and the undesired p-isomer. Corey employed two different principles to obtain the desired cis configuration at C-8 and C-9: ketene addition to a double bond41(Fig. 9) and iodoiactone formations7 (Fig. 10). (Prostaglandin numbering is used for synthetic intermediates throughout this paper.) Both approaches not only assured the correct stereochemistry and functionality 0
0 B
B
OCH,
+
OCH,
I
I I
H
H
Figure 9
0 7
6
Figure 10
--+
92
The Total Synthesis of Prostaglandins
<
Figure 11
on the five-membered ring but also allowed a convenient introduction of the A5 double bond in the synthesis of PGF,a: Baeyer-Villiger oxidation of the cyclobutanone produced a lactone similar to the one in Fig. 10, which was reduced to a hemiacetal and then subjected to a Wittig reaction (Fig. 11). The Wittig condensation was carried out under conditions which were known to produce selectively ~is-olefins.~~ This strategy, construction of a y-lactone fused to the cyclopentane ring at C-8 and C-9, proved to be very efficient since it allowed the introduction of the right functionalities with correct stereochemistry at three centers. Introduction of the carboxy side chain by direct alkylation of the 9-ketone resulted in mixtures of h a n d 8 # ? - i s o m e r ~ ~(Fig. ~ . ~ 12). l - ~The ~ ~ thermodynamic. equilibrium was found to be 65:35 in favor of the 8#?-isomerwhen exo configuration at C-13 and 80:20 in favor of the 8a-isomer in the endo series. These results were surprising at first since the base-catalyzed equilibrium of PGE, and 8-iso-PGE1 gave a 9O:lO mixture in favor of the natural 8aconfiguration, which was expected because trans configuration of the two side chains should be favored thermodynamically. The apparent preference
8a
13
88 Figure 12
2. Synthetic Routes to Prostaglandis E and F
93
for cis substitution at C-8 and C-12 in the em-bicyclohexane system is probably due to a boatlike conformation of the five-membered ring in bicyclo[3.I.O]hexanes (see, e.g., citations 8 and 13 in reference 71). In such a conformation the 8S-oriented side chain assumes an equatorial position. In the endo-bicyclohexane system, however, steric interaction with the substituent at (2-13 causes preference of the 8a-position. In other syntheses the more stable trans configuration at C-8 and C-12 is obtained by thermodynamically controlled formation of the five-membered ring3s*3e-38*45 or during generation of the asymmetric center at C-12 under basic conditions.116
CN Figure 13
Stereochemical control at C-11 and C-12 proved to be not too difficult. In one of Corey's syntheses Diels-Alder addition assures the correct trans-trans relationship at Ca-C1,-C,, (Fig. 13).37When the hydroxyl-group at C-11 was introduced by opening of a cyclopropyl ring only the 110: isomer was formed14.41.70.71,105~10e (Fig. 14). Whether this is due to mechanistic control, that is, the approach of the incoming hydroxyl group from the opposite side, or product control, the formation of the more stable trans configuration, depends on how much carbonium ion character one is willing to concede C-11 during opening of the cyclopropane ring. +
13
Figure 14
OH
11
13
In the Merck synthesis trans relationship between substituents at C-I1 and C-12 was achieved by base-catalyzed epimerization of an 1I-acetyl group.116 Where the five-membered ring was formed via aldol condensation originally a mixture of 11-epimers was ~ b t a i n e d ~but ~ * later ~ " conditions were found which gave exclusively the desired a-isomer (Fig. 15).38Surprisingly, even yielded mainly the direct epoxidation of 2-oxabicyclo[3.3.0]-oct-6-en-3-one desired a-epoxide (Fig. 16).42 Two methods were used for formation of the allyl-alcohol system: (1) generation ofan enone unit followed by reduction of the k e t o g r o ~ p ~ ~ - ~ ~ ~ ~ ~ (Fig. 17); and (2) solvolysis of a bicyclo[3.1.O]-hexane ~ y ~ t e m ~ ~ * ~ (Fig. 18). In both cases only the desired trans olefin is formed. In the first
94
The Total Synthesis of Prostaglandins SnCI,
___t
acetone
HO
0
Figure 15
Figure 16
Figure 17
method trans geometry is assured because of the readily obtainable transenone; in the second method the activation energy for the rearrangement probably is high enough so that the more stable rrans compound is formed without stopping at the energetically less stable cis olefin. The configuration at C-14 (Fig. 18) has no influence on stereochemistry or yield of the solvolysis reaction; both epimers give the same products. Unfortunately both methods result in mixtures of C-15 epimers, in the case of the latter again regardless of configuration at C-15 before solvolysis; so far in all prostaglandin syntheses the C-15 epimers have to be separated by column chromatography. The only exception has recently been disclosed by Corey;46 in this variation a phosphonium salt with the correct 15(S) configuration is prepared in a five step synthesis from L(-) malic acid and then reacted with an aldehyde to afford exclusively the desired 15a-isomer (Fig. 19).
Figure 18
z
a
0 )
4
:;a z
0
h
d
OX
z
c
E t(
95
96
The Total Synthesis of Prostaglandins
Figure 20
As mentioned, Corey et al. introduced the A5-cis double bond in PGE, and PGF,cc by Wittig reaction under specific conditions. Another possibility was realized by Schneider,lo4who obtained the required cis geometry by hydrogenation of an acetylenic bond (Fig. 20). The same principle was utilized in the total synthesis of d,l-PGE, methylester, where both cis double bonds were generated simultaneously by hydrogenation of an acetylenic precursor (Fig. 21).15
0
Figure 21
C. Individual Syntheses How general strategy and stereochemical principles were implemented using the most efficient reaction sequences and best suitable protecting groups is outlined in the following charts. The figures describe the total syntheses of PGE’s and PGF’s completely, including the reagents used; we therefore limit the discussion, to pointing out the highlights, advantages, or disadvantages of the different syntheses. Corey’s early s y n t h e ~ e s 3 ~ ~have ~ ~ - ~in~common ~ Q O that the five-membered ring is formed by aldol condensation during the course of the synthesis and that the 9-keto group is generated at the end from an amine precursor (Figs. 22-25). The first synthesis,3b leading to d,l-PGEI, is outlined in Fig. 22. Noteworthy is the stereospecific Diels-Alder addition which gave adduct 4 as the major product and the position isomer only as a minor by-product. Base-catalyzed aldol cyclization of 9 yielded in addition to 10 its 11-epimer.
1
2
3
4
6
7
Figure 22 97
19
d, I-PGE, Figure 22 (contd.)
98
2. Synthetic Routes to Prostaglandis E and F
99
Important for the development of alater synthesis (Figs. 31,32) was the finding that the tetrahydropyranyl-group at C-1 1 and C-15 could be removed under conditions which would not destroy the acid-sensitive 8-ketol unit. Another method3e to prepare intermediate 14 (Fig. 23) involved acidcatalyzed aldol condensation. It was found that the A13-double bond in 23 had to be transposed to the A12-positionfor the cyclization; otherwise useless reaction products were obtained.40Using p-toluene sulfonic acid as catalyst, again a mixture of 11-epimers was obtained. The formation of the undesired 11-epimer could be suppressed, however, by using stannic chloride as catalyst for the cyclization of intermediate 24. In this variations8 (Fig. 24) amine 28 was prepared and the amino group at C-9 could then be used as an handle for resolution via the a-bromocamphor.rr-sulfonate. Conversion of resolved amine 28 by the procedures developed earlier (e.g., Fig. 22) led to the first total synthesis of resolved PGE,. A very short route to intermediate 25 (Fig. 25) was disclosed recently by C ~ r e y ; ~ ~ unfortunately, this route is limited by low yield in the Wittig reaction (31 -+ 32). Another approach, based on a concept by Just and Simonovitch70 and developed by the Upjohn involves solvolysis of a bicyclo [3.1.O]-hexane system as its key step. Intermediate 45 was prepared from A3-cyclopentenol (Fig. 26). The original concept ,70 formolysis of epoxide 47, was successful only in the PGF-series and only in very low yield.?l Preparation of bismesylate 49 (Fig. 27) followed by solvolysis in acetone/ water allowed adaption of this route to the synthesis of d,EPGE1 methylester and d,l-PGE1.106*106The yield of solvolysis products was in the order of 10% and was found to be independent of the configuration at (2-14 and C-15 in bismesylate 49. Minor isomers, however, isolated from this route and identified as C-13 isomers gave considerably higher yields of solvolysis products, which led to the development of a new synthesis, designed to produce only these endo isomers14 (Fig. 28). Solvolysis of endo-bismesylates 65 gave solvolysis products in better than 40% yield. The higher yield in the endo series is probably due to steric hindrance at C-14, which makes the competing reaction, straight hydrolysis at C-14 without ring opening, less predominant than in the exo series. Modifications of the endo-bicyclohexane synthesis allowed the first total syntheses of d,l-PGE,'04 (Fig. 29) and d,lPGE, methylester'6 (Fig. 30). This approach, although adaptable to the synthesis of all prostaglandins known so far and a wide variety of analogous compounds, lacks stereocontrol except for the solvolysis reaction. A stereocontrolled synthesis of PGF,a by Corey et a1.37*43*a is characterized by the build-up of a y-lactone fused to the five-membered ring at C-8 and C-9, which is used for generation of the 9a-hydroxyl group and the A6-cisdouble bond (Fig. 31). Noteworthy is the selective Baeyer-Villiger oxidation of ketone
/cH \ \CHO
CH30
OCHa 22
(CH 2 16CN
OH
0
26 AcpO, Pyr
100
Figure 23
14
27
Figure 24
28
101
45 Figure 26
102
46
CH,SO,CI
&;6c02cH3
HpOlacclonc I4
I///,
49
-
-
15
CH-CH-C,H,,
I
&-3:E3
I
HO
-
OH
d.1-PGE, methyl ester I 5-epimer
OMS OMS
+
Figure 27 103
52
53
54
65
55
d,l-PGE, methyl ester Figure 28
104
0 N
8:."
0
O U '-
7
X
U
0
43'; 105
z
0 0 V
q
I
4
+
106
g,. 81
\\I\\\\"-
=
-
CO,C H
-\
+ 15-epi iiicr
6 I4
HO
82
'*
Hs, Lindlar Quinoline
' HO
OH
d,l-PGE, methyl ester Figure 30 107
C'H,0CH3
?I-CI
I
CHz-C-CN
84
83
CH,OC'H,
&c{H30cH&
85
P m-CI-C,.H,C03H
CN
86
0
CH30CH2
&NaOll
87
108
0
QC
H 2 CO 0H
H e w l u inn ~ _____, ephcdrinc via
otr
CHZOCH3 88
\.ill
0
50 95
0
98
OH J
99
Figure 31 (contd.) 109
110
The Total Synthesis of Prostaglandins
86. Acid 88 was readily resolved via the ephedrine salt and iodolactonization of resolved acid 89 afforded key intermediate 90 with correct stereochemistry at all four asymmetric centers of the ring. The earlier finding36 that 11,15ditetrahydropyranyl ethers of PGE compounds can be hydrolyzed without B-elimination of the 11-hydroxy group permitted the synthesis of PGE, from intermediate 101 (Fig. 32). The same intermediate could also be selectively hydrogenated to give PGE, and PGF,a (Fig. 32). A shorter approach4' to intermediate 95 is handicapped by a very low yield in the ring-opening step 109 to 110 (Fig. 33). A third synthesis of PGF,a42 using the y-lactone principle is short and direct but lacks stereospecificity in the opening of epoxide 117 (Fig. 34). A unique, if lengthy, synthesis of d,I-PGE, has recently been published by a group at M e r ~ k . "The ~ intention was to exercise stereochemical control at all asymmetric centers (except (2-15) and this has been achieved, although at the cost of additional steps; for example, the double bond in intermediate 136 is only hydrogenated after shift to the AlO-positionunder basic conditions to assure trans configuration at C-8 and C-12 (Fig. 35). The correct stereochemistry at C-11 is obtained in a similar manner by epimerization of intermediate 143 to the thermodynamically more stable isomer 144. Several ofthe syntheses outlined here provide practical routes to the pharmacologically important E and F prostaglandins and in several instances this has been achieved by developing new methodology which may be useful for other tasks.
3. PARTIAL SYNTHESES OF PCE, AND PCF,a FROM lS(R)-PGA,
In 1969, Weinheimer and Spraggins r e p ~ r t e d " the ~ isolation of surprisingly large amounts of lS(R)-PGA, (154) and its 15-acetate, methyl ester (155) from a sea whip, Plexaura /iomomalla, a common gorgonian coral of Florida coastal waters. The combined weights of these two prostaglandins represented about 1.5% of the dry weight of the coral cortex. Before this finding, the highest concentration of prostaglandins found in nature was in mammalian semen, which would present.some problems if considered as a source of large amounts of prostaglandins. These prostaglandins, 154-156, the latter the result of partial hydrolysis of 155, are not presently of major biological interest. They have, however, been converted3, to the highly active prostaglandins, PGF,a and PGE, methyl ester. The two major problems in such a conversion involve inversion of configuration at C-15 and, formally, hydration of the 10,ll-double bond. The second was accomplished by epoxidation with hydrogen peroxide and a small amount of potassium hydroxide in methanol, affording a mixture of epoxides. The ratio of a to epoxide formed was 75:25 from 155.
101
-
AcOH/H20/THF*
Joncr
OTHP
&-~p 102
6H
6H PGE2
HQ H I'd--C
AcOlI/H O/THF 2
101 -&p
OTHP
--
6THp 103
HO
c
(3H
---
OH
PGF,a AcOH/ H201THF
$
OTHP
&-HP 104
OH
6H PGE,
Figure 32 111
Zn
__*
108
107
0
BK"
0
,A H ;/;
109
H
0,
3 OH
C,HnCCH2P(OCH,I2
f
CH0
b % I10
0-4
0-u"
AclO/Pyr
,
3
OH
0
OAC
111
Figure 33
112
0 95
114 115
116
0TH3
+ ?(
SCH,
""0
+
SCH,
OH
117
119
SCH, 118
HgCIz, CaCOa CHsCN, H z 0
posit ion
+ isomer .
c oqcH3 \, '
nCsHIILi ____*
CHO
-
OH 120
oxC" + 15-epimer OH
--
122
OH 121
Figure 34 113
$
16:
'Ph3Br-
PhsP.HBrb
OHC(CH&.CO&Ha,
OCH3
OCH, 124
125
%
C H (CH*),C02CH3 C P ,COOH
OCH,
qj
126
(CH ),C02CH3
(CH, ) ,CO,CH
OCH 127
=$$H
___* P-TsOH
OCH3
128
Figure 35 114
P
I
OCHB
OCH3
129
130
OCH3 131
132
HOAc
OCH3
133
Miel, Ph&Li
0
134
LiAIH(0Bu‘h
135
one
4 10
H3C
(CH2),,C02CH3
1’’
HCIO,
OH 136
Figure 35 (contd.)
115
i ,H
139
I50
151
117
118
The Total Synthesis of Prostaglandins
Reductive opening of the epoxide mixture by chromous acetate in aqueous acetic gave IS(R)-PGE,, 15-acetate, methyl ester (159), which was separated chromatographically from the 1 16-isomer. This was converted to its trimethylsilyl ether and reduced with sodium borohydride to give 9ahydroxy compound 160 (55 % yield from 159), separated chromatographically from the corresponding 9/?-isomer. The 9a:9/? ratio was considerably less if the silylation step was omitted. The purified 160 was hydrolyzed to 161 and selectively oxidized with 2,3-dichloro-5,6 dicyano-1 ,Cbenzoquinone in dioxane to 162. Reduction of the 15-ketone with zinc borohydride in dirnetho~yethane~'gave a mixture of PGF,a (163) and its 15-epimer. Again, epimer ratio was more favorable [73:27 = 15(S):15(R)] when the hydroxyl groups were first protected as trimethylsilyl ethers. A modification of the preceding route allowed the preparation of PGE, methyl ester from the mono ester 156. Inversion at C,, was accomplished in fair yield by forming the 15-methanesulfonate and solvolyzing it in acetonewater. The desired product 164 was accompanied by starting material (156) -and several by-products. The introduction of the 11a-hydroxyl group into 164 was accomplished as in the previous example; epoxidation to 165 and chromous acetate reduction gave a mixture of 11-epimers from which PGE, methyl ester (166) was separated by chromatography.
LmCOOR
R'O
"H
154 R = R' = H
CH,, R' R' = AC H 155 156 R = CH,,
0 0 ' " . 0
AcO
'H
157 a-epoxide 158 P-epoxide
HO G
O
O
R f-
~ i j R’O
160 R = CH,, R‘ = AC 161 R = R ’ = H
166 119
120
The Total Synthesis of Prostaglandins
4. SYNTHETIC ROUTES TO STRUCTURALLY SIMPLIFIED PROSTANOIC ACIDS
The restraints placed upon available synthetic methods by the variety of functional groups of natural prostaglandins are, of course, lessened by making simplified analogs the synthetic goal. For example, the preparation of completely saturated prostaglandins allows one to use a wide range of reducing agents not possible if PGEa is the objective of synthesis. In some of the following syntheses, stereochemical control has also been sacrificed; isomers were not always separated, making evaluation of the work by biological or other activity difficult. However, a number of these syntheses make use of novel reactions or unusual protective groups and so have intrinsic value for synthetic organic chemistry.
A. d,l-13,14-Dihydro-PGEl 13,14-Dihydro-PGE1is a naturally occurring, biologically active metabolite of PGE,.' The synthesis, in 1966, of the racemic ethyl ester 172 of this material by Beal, Babcock, and L i n ~ o l represents n ~ ~ ~ ~ the ~ first synthesis of a biologically active natural prostaglandin. Their route illustrates a novel use of the Wittig olefination reaction and represents an extremely short and efficient synthesis of the prostanoic acid skeleton. The sodium enolate of 5-formyl-3-ethoxy-2-cyclopentenone167 was reacted with the phosphonium bromide 168 derived from ethyl 6-bromosorbate and triphenylphosphine to give 169, via the in situ formed ylid. Catalylic hydrogenation and reformylation was followed by a second, similar Wittig reaction, this time using n-hexanoylmethyltriphenylphosphonium chloride 173 to give 170. In this product, the prostaglandin skeleton and oxygen substitution pattern is already established. A series of reductive steps, after changing ethyl enol ether to benzyl enol ether, gave 172 as a mixture of stereoisomers. Chromatographic separation gave material having biological activity and the same polarity and spectral properties as authentic natural dihydro-PGE, ethyl ester. In addition, an isotope dilution method was used to demonstrate the presence of at least 22% of the natural isomer. This synthesis suffers from a low yield only in the final reduction of the benzyl enol ether system to the P-hydroxy-ketone 171 172. Klok, Pabon, and van DorpT4have also reported a synthesis of a mixture of steroisomers of dihydro-PGE, as an extension of their synthesis of d,fPGB, and d,l-PGE1-237 (174) (sec below). After acetylation of the hydroxyl group of 174, allylic bromination followed by displacement of bromide with silver acetate gave 175. The ester functions were hydrolyzed and --f
Irfi
*
%
0
0
' ew
121
122
The Total Synthesis of Prostaglandins
the resulting 176 reduced over rhodium-charcoal catalyst to give a mixture of 177 and 178, the latter unfortunately predominating. The diol 177 was separated by chromatography and was shown to be spectrally identical to 13,14-dihydro-PGE1. It had about 15% of the biological activity of the natural material.
%OH
%OH+
OH
RO
OR 175 K = A c 176 K = t I
174
1
%OH+
OH 178
c'cr^,u--
HO
OH 171
A third synthesis of the same PGE, metabolite, again as a mixture of stereoisomers, has recently been reported by Strike and Smith.l13A series of alkylation steps was used to build up the acyclic structure 179.After removal of the 1-butoxycarbonyl group, the acetylenic side chain was partially reduced to give 180.Ozonolysis converted this chain to an aldehyde which underwent aldolization with base to 181. None of the desired intermediate 8-hydroxyketone was obtained. Epoxidation (alkaline hydrogen peroxide), catalytic hydrogenation over palladium-charcoal, and acidic removal of the tetrahydropyranyl group gave a stereoisomeric mixture having spectral and chromatographic properties consistent with structure 183.It was not shown how much of the natural stereoisomer was present in the mixture, which did have considerable biological activity.
B. 15-Dehydro-PGE1 Miyano and D ~ r n carried '~ out a short and efficient synthesis of a mixture of stereoisomers said to be spectrally indistinguishable from 1 5-dehydro-PGE1 (1 87). Condensation of 3-keto-undecan-l ,l l-dioic acid with styrylglyoxal gave crystalline 184,which was cyclized to the cyclopentenone 185. Cleavage
8: X 0 0
G
0
t
8 \
0
X
123
124
The Total Synthesis of Prostaglandins
at the styryl double bond gave aldehyde 186 in which the conjugated double bond was then reduced with zinc. A Wittig olefination with n-hexanoylmethylenetriphenylphosphorane gave a mixture of stereoisomers of structure 187. Evidently selective reduction of the 15-ketone to PGE, was unsuccessful. Borohydride reduction of both keto groups to PGF,a was not mentioned, but this would have presumably given proof of structure and stereochemistry of their product.
icvvv.ooH -
I86
185
1
187
C.
11-Desoxy Prostaglandins
Several 1 1-desoxy prostaglandins were prepared by Bagli and Bogri16 of Ayerst in 1966 and these unnatural materials were shown to retain prostaglandin-like activity. These authors ~ubsequently'~ improved the synthesis and proved the stereochemical structure of their products. In the later work, the substituted cyclopentenone 189 was prepared by alkylation of ethyl 2-cyclopentanone carboxylate (188), bromination, and treatment with ethanolic sulfuric acid. Addition of HCN to 189 and subsequent hydrolysis gave keto acid 190, which was monoesterifieh with p-toluenesulfonic acid in methanol (191). The acid was converted to its acid chloride 193, which added
I 0
h
2 -
ij
I
I_ 0 0
Li
rrr
f
t
0 0
b
3
125
126
The Total Synthesis of Prostaglandins
to I-heptyne in the presence of aluminum chloride to give 194. This chlorovinyl ketone was converted with methanolic sodium hydroxide to enol ether 195. Sodium borohydride reduction of both ketones and acidification gave the 13,14-unsaturated-l5-ketone196. Further borohydride reduction produced d,l-11-desoxy -PGF,/? and its 15-epimer 197, which were not separated. Presumably d,l-I l-desoxy-F,a 198 was also produced as a minor product, since these authors noted that borohydride reduction of the diester 19 gave two isomeric 9-alcohols, 9a: 9j3 = 15 :85. In view of the later observations of Ramwell et al.,g6it might be surmised that much of the biological activity mentioned by Bagli et al. for the 9b-01 197 may be due to the ent-1 I-desoxy15-eyi-PGFlj3 and/or 1 l-desoxy-F,a present in the mixture, rather than to nat- 1 l-desoxy-F,p. A novel approach to prostanoic acids via a photochemically prepared bicyclo[3,2,0] heptanone has also been used by Bagli and Bogri.la The same substituted cyclopentenone 189 used above was irradiated (high-pressure mercury lamp, Pyrex filter) with the chlorovinyl ketone 199 to give adduct 200. When this photoadduct was refluxed with zinc in acetic acid, the major product isolated (44%) was the methyl prostanoate diketone 201. The minor product was simply dechlorinated adduct. Sodium borohydride reduction of the diketone gave a mixture of diols which was separated into two fractions by chromatography, reported to be the 9-epimers 202 and 203, both of which had biological activity.
D. Cyclopentenone Prostanoic Acids The substituted cyclopentenone 207 [the 15(S)-epimer], or prostaglandin B,, is the end-product of a series of base catalyzed steps involving dehydration of PGE, and double-bond r n i g r a t i o n ~Racemic .~~ and also optically active PGB, have been synthesized by a number of groups, not all of which had this product as their initial goal. These syntheses give some valuable methods for cyclopentene substitution. The first reported synthesis of d,l-PGB, was by Hardegger and associates.66 The substituted cyclopentenone 189 was condensed with 3-t-butoxyoctynyl magnesium bromide. Acid-catalyzed allylic rearrangement of the initial product 204 and oxidation gave the disubstituted cyclopentenone 206. Partial catalytic reduction of the triple bond, removal of the r-butyl ether, and base hydrolysis completed the synthesis of d,l-PGB1 207. The basic hydrolysis step also served to isomerize the initially cis-13,14 double bond to trans. Complete reduction of the triple bond of 206 and removal of protective groupsgave the racemic form of PGE-237 (208), also previously obtained from PGE,.89A very similar synthesis of 207 and 208 was also reported by Klok, Pabon, and van D ~ r p . ~ ~
--x
:
t a
3
127
t
128
4. Synthetic Routes to Structurally Simplifted Prostanoic Acids
129
Three other g r o ~ p s ~prepared ~ * ~ ~PGB, * ~ from ~ ~ the 2-substituted cyclopentan-l,3-dione en01 ether 209. This starting material was prepared by Collins, Jung, and Pappo from the hydroxydione 212, acid-catalyzed hydrogenolysis and en01 ether formation of which gave 209. Yura and Ide cyclized the keto acid 213 using aluminum chloride and propionyl chloride at 80". Katsube and Matsui prepared 209 from ethyl 9-oxodecanoate. All three groups then reacted 209 with the acetylenic magnesium bromide from
+
coo"-
W209R-H or O Et
0
H
O
C--C-CH--C,H,,
I
210
OTHP
R
W
O
O
R
C=C--CHC,HlI
I
OH
211
1-octyn-3-01, tetrahydropyranyl ether to give adduct 210. Acid hydrolysis of the en01 ether and concomitant dehydration gave the cyclopentenone 211 which was converted to PGB, as in the preceding route. Pappoe2 recently reported the resolution of the intermediate 1-octyn-3-01 and the use of the (S)-enantiomer for the synthesis of natural lS(S)-PGB,. The method of Dale, Dull, and MosherdsR involving an N M R analysis of the derived (-)a-methoxy-a-trifluoromethylphenyl acetate ester was used to determine the enantiomeric purity of the resolved 1-octyn-3-01s. This method has also been used107 to assay mixtures of 15(R) and 15(S)-PGB2. In the same report,Q2Pappo described the conversion of the triketo acid 214 to the enol ether 215. This, with difficulty, condensed with the same octynyl magnesium bromide to give 216, which is the 11-hydroxy analog of 211, opening the way to synthesis of 11-hydroxy prostaglandins. In this connection, Vandewalle, Sipido, and DeWilde116carried out some similar studies with the simpler triketone 217. They converted it in a series of steps to 218, which on lithium-ammonia rcduction gave some 219, stereocheniistry
WOOH wo The Total Synthesis of Prostaglandins
130
__c
0
HO
214
K3
21s
0
HO
217
1
216
H0 &GH,
HO
219
OH
unspecified, among other products. This is being investigated further as a potential PGF,a synthesis. An attempt to repeat the Just PGE, synthesis by Holden et aLaoagave d,l-PGB, instead. Morin et al.81 also describe a synthesis via the aromatic intermediate 220. which gave a poorly characterized mixture said to contain 221 and the methyl ester of d,l-PGB,. WOCH,
O
O
H
k
o
H
I
220 221
OH
A synthesis of the 15-keto analog of racemic PGB, methyl ester by M i y a n ~ ~illustrates ~ * * ~ the unusual use of a bicyclo[2,2,1]heptene as a protecting group and precursor of a trans carbon-carbon double bond. The starting material 222, prepared from 5- norbornene-2,3-endodicarboxylic anhydride, was condensed with the sodio-enolate of dimethyl 3-oxoundecan1.1 1-dioate to afford the triketo diester 223. This was cyclized to the diketo
4. Synthetic Routes to Structurally Simplified Prostanoic Acids
131
diacid 224 with base. Decarboxylation (copper-quinoline) and esterification gave 225, which was pyrolyzed, producing 1 5-dehydro-PGB1 methyl ester 226 by a reverse Diels-Alder reaction. The new double bond was shown to be 0
CHSOOC
223
222
FocH3
1
t-
+
HOOC
-
225
224
227
226
OH
228
trans, and could be selectively reduced with zinc in acetic acid to give 227,
which is 15-dehydro PGE-237. Some d,l-PGE-237 (228) was obtained by catalytic hydrogenation of 226, along with 227. The earliest synthesis of a cyclopentenone prostanoic acid was by Samuelsson and Stallberg.QaThe two routes shown lead to a chromatographically separable mixture of 229 and 230, the latter being identical to a degradation product of PGE,.QQ
132
The Total Synthesis of Prostaglandins
0
0
I1
CH,OOC-(CH,),C-CH,COOCH, 0
II
+ Br-CH,C-(CH,),CH,
0
II
I1
CH,OOC-(CHJ,C-CH-CH,C-(CHJJH,
I
COOCH,
229
2 II
CH,OOC(CH,),C-CH,CH,COOCH,
230
+ C,H,,MgBr
E. d,l-PGE, Methoxime In a synthesis involving cyclopentenone intermediates similar to some of those in Section 4.D, Finch and F i P have succeeded in preparing d,l-PGE1 methoxime. Although in a simple model system the methoxime of a @hydroxyketone could be hydrolyzed to the 8-hydroxyketone, these authors were unable to isolate pure d,l-PGE, from hydrolysis attempts from their synthetic methoxime. They mention a new oxime reagent they developed which can be removed under very mild conditions, but this has not yet been disclosed. The cyclopentenone diester 231 was allylically brominated and the resulting bromide displaced by silver acetate to give acetate dimethyl ester 232.
4. Synthetic Routes to Structurally Simplifted Prostanoic Acids
133
Methanolysis of the acetate gave alcohol 233, which was silylated and then reduced with hydrogen on Raney nickel. Silylation was said to direct the cishydrogenation to the side of the ring opposite the bulky silyloxy group, producing the all cis stereochemistry as shown in 234, Formation of the 9methoxime and hydrolysis of the silyl ether gave 235,which was much more resistant to dehydration than the corresponding /?-hydroketone. Saponification (methanolic K,CO,) and reesterification gave the new diester 236 in which epimerization a to the carbomethoxy group was assumed to have taken place. Epimerization a to the methoxime was considered unlikely based on model studies. The tetrahydropyranyl ether of 236 was prepared and treated with sodium borohydride in ethanol, giving alcohol 237 (40% yield). It was thought that complexing of borohydride with the tetrahydropyranyl group (or methoxime?) helped to give the desired selective reduction. Oxidation of alcohol 237 to the aldehyde and Wittig olefination with tributylphosphoranylidene-Zheptanone gave unsaturated ketone 238. Borohydride reduction of 238 followed by hydrolysis of tetrahydropyranyl ether and methyl ester produced a mixture of diol acids from which a crystalline isomer 239 (m.p. 97-99') was separated. This racemic material was compared with one of the isomers (syn or anti?) of the methoxime (m.p. 55-57") of natural PGE, by the usual criteria, but a satisfactory hydrolysis of this to d,l-PGE, was not achieved. 0 0 COOCH3
PCOOCH,
23-~1
234
1
235
-
AcO
232
I
233
236
134
The Total Synthesis of Prostaglandins
236
1
:
H6
‘0H 239
5. OXAPROSTAGLANDINS
Fried and co-workerss6 have carried out a well-planned synthesis of ’I-oxaprostaglandins in which the trans opening of an epoxide is used to insure the correct steric relationships of the four substituents on the cyclopentane ring. The all cis-l,2-epoxycyclopentan-3,5-diol240 was dibenzylated and the resulting diether 241 reacted with diethyl-1-octynylalane, giving 242. The dialkyl alkynyl aluminum reagent gives nearly quantitative yields of 242 at room temperature, contrasting with the failure of alkali or magnesium acetylides to react even at elevated temperatures. The alane can be generated in situ30and employed directly in toluene solution. The resulting alcohol was then alkylated with Et,AI
t
NEt,
+ HCEC-R
/c=-c-R
+ EtzAI
+ CZH,
\Ets
t-butyl 6-iodohexanoate (with dimsyl sodium in DMSO) to give ester 243 in 65 % yield. Removal of the 1-butyl group with trifluoroacetic acid gave acid
5.
Oxaprostaglandins
135
243 R = t-butyl 244 R = H
246
245
244 which was reduced with lithium in methylamine. This removed the benzyl groups as well as reducing the acetylenic bond to a trans double bond, producing 245. The synthesis of 7-oxa-PGF1a (246) was completed by selenium dioxide oxidation, which was nonstereospecific. It was found subsequentlyso that intermediate 242 could be resolved via reaction with (+) or (-)-cr-phenethyl isocyanate and recrystallization of the urethanes. The absolute configuration of the resolved (+) 242 was determined by catalytic hydrogenation to triol 247, acetonide formation (248) and oxidation to ketone 249. Optical rotatory dispersion studies of this ketone were definitive in assigning configuration. The ability to form acetonides from such oic-glycols was used to accomplish a synthesis of 7-oxa-PGE,. The required triol 251 was obtained from 250 by reaction as above with an acetylenic alane and hydrolysis of silyl groups. Now, acetonide formation (252) and benzylation gave 253, which could be
247
248
249
(CH3)3 SQ
(cII,)~s~~ 250
255 256
136
R = t-butyl R =H
251
J
252 253
254
R =H R =CH,+
5. Oxaprostaglandins
137
selectively hydrolyzed (aqueous trifluoroacetic acid) to glycol 254. Alkylation as above unexpectedly gave 65% of the desired product 255 and only 20% of the other possible monoalkylated product. Hydrolysis of the t-butyl ester and oxidation gave keto-acid 257. The keto group was protected from reduction by ketal formation, which simultaneously esterified the acid. Saponification of 258 followed by lithium-methylamine reduction gave 259, which was allylically hydroxylated with selenium dioxide. Removal of the ketal group with trifluoroacetic acid at 0” gave 7-oxa-PGE, (260), found to be sensitive to dehydration of the /I-hydroxy-ketone system. Application of similar synthetic principles by Fried has led to a number of other oxa-prostanoic acids.60A further extension is proposed as a synthesis of PGF,a. The isomeric “trans” expoxy dibenzyl ether 261 was used as starting material. Reaction as before with an acetylenic alane reagent gave 262, where now the acetylenic side chain is to be used as the precursor of the carboxylic acid side chain of the prostanoic acid. The derived diol263 could be resolved via a diurethane from (+)-a-phenethyl isocyanate. Tritylation of the primary hydroxyl and tosylation of the secondary gave 266, which was simultaneously debenzylated, detritylated, and the triple bond saturated by hydrogenation over palladium catalyst. Base treatment of the resulting 265 resulted in epoxide formation 266. Silylation of the diol and reaction with another acetylenic alane gave a mixture of two isomeric products, 267 and 269. The desired prostanol267, which is unfortunately the minor product of the epoxide opening, was reduced to 270. Further proposed transformations of 270 involving protection of secondary alcohols and oxidation of the primary alcohol should give PGF,ar.
261
262 R = THP
263
R =H
264
The Total Synthesis of Prostaglandins
138
264
,
P O T S HCj
I
"0
266
265
OR
267 R = THP 268 R = H
269
1
REFERENCES 1. S. Abrahamsson, S. Bergstrom, and B. Samuelsson. Proc. chem. Soc., 1962, 332. 2. S.Abrahamsson. ACIUCrystullog., 16,409 (1963). 3. D.G. Ahern and D. T. Downing. Biochini. Biopliys. ACIU,210, 456 (1970). 4. D. G. Ahern and D. T. Downing. Fed. Proc. Fed. Am. Socs. exp. Biol., 29, 854 (1 970). (Abstract). 5. N. H. Andersen. 1. Lipid Res., 10, 316 (1969). 6. N. H. Andersen. J . Lipid Res., 10, 320 (1969). 7. E.AnggArd and B. Samuelsson. J . Biol. Cheni., 239,4097 (1964). 8. E. AnggArd and B. Samuelsson. J . Biol. Chem., 240,3518 (1965). 9. E. AnggArd and B. Samuelsson. Ark. Kemi, 25,293 (1966). 10. E. Angghrd and B. Samuelsson. Mem. SOC.Endocr., 14, 107 (1966). 11. E. AnggArd and B. Samuelsson. ActuPhysiol. Scund., 68 (Suppl. 277), 232 (1966). 12. E.AnggArd and B. Samuelsson. Meth. Enzyniol., 14,215 (1969).
References
139
13. U. Axen. In C. K. Cain, Ed., AnniialReportsiir Medicinal Chemistry 1967, Academic Press, New York, 1968. pp. 290-296. 14. U. Axen, F. H. Lincoln, and J. L. Thompson. Cherrt. Cornr~wr.,1969, 303. 15. U.Axen, J. L. Thompson, and 3. E. Pike. C/iem. Commun.,1970,602. 16. J. F.Bagli, T. Bogri, R. Deghenghi, and K. Wiesner. Tetrahedron Lett., 1966, 465. 17. J. F.Bagli and T. Bogri. Tetrahedron Lett., 1967, 5. 18. J. F.Bagli and T. Bogri. Tetrahedron Lett., 1969, 1639. 19. 3. F. Bagli and T. Bogri. Abstracts, 158th Meeting Am. Chem. SOC.,New York, 8-12 September 1969, MEDI 4. 20. J. F. Bagli. In C. K. Cain, Ed., Annual Reports in Medicinal Cheniistry, Academic Press, New York, 1970,pp. 170-179. 21. J. F. Bagli and T. Bogri. Abstracts, 5th Middle Atlantic Regional Meeting, Am. Chem. SOC.,Newark, Del., 1-3 April 1970,p. 59. 22. P. F.Beal, 111, J. C. Babcock, and F. H. Linco1n.J. Am, Chem.SOC.,88,3131 (1966). 23. P. F.Beal, J. C. Babcock, and F. H. Lincoln. In S. Bergstrom and B. Samuelsson, Eds., Nobel Symposium 2, Prostaglandins, Almqvist and Wiksell, Stockholm, 1967,pp. 219-230. 24. S. Bergstrom, L. Krabisch, B. Samuelsson, and J. Sjovall. Acta. Clreni. Scatid., 16, 969 (1962). 25. S. Bergstrom, R. Ryhage, B. Samuelsson, and J. SJSvall. J . Biol. Chem., 238, 3555 (1963). 26. S. Bergstrom, H. Danielsson, D. Klenberg, and B. Samuelsson. J. Biol. Chem,, 239, PC4006 (1964). 27. S. Bergstrom, H.Danielsson, and B. Samuelsson. Biochim. Biophys. Acta, 90, 207 (1964). 28. S. Bergstram. Science, 157, 382 (1967). 29. S. Bergstrom, L. A. Carlson, and J. R. Weeks. Pharmac. Rev., 20, 1 (1968). 30. P.Binger. Angew. chem., 75,918 (1963). 31. G.L. Bundy, F. H. Lincoln, N. A. Nelson, J. E. Pike, and W. P. Schneider. Ann. N.Y.Acad. Sci., 180, 76 (1971). 32. W.Cole and P. J. Julian. J. Org. Chem., 19, 131 (1954). 33. A. Collet and J. Jacques. Chitti. /her., 5 , 163 (1970). 34. P. Collins, C. J. Jung, and R. Pappo. IsraelJ. Chenr., 6,839 (1968). 35. E. J. Corey, N. H. Andersen. R. M.Carlson, J. Paust, E. Vedejs, I. Vlattas, and R. E. K . Winter. J. Am. Chenr. Soc., 90, 3245 (1968). 36. E.J. Corey, I. Vlattas, N. H. Andersen, and K . Harding. J. Am. Cheni. SOC.,90, 3247 (1968) [see erratum 90,5947 (1968)l. 37. E. J. Corey, N. M. Weinshenker,T. K. Schaaf. and W. Huber.J. Am. Chenr. SOC..91, 5675 (1969). 38. E. J. Corey, I. Vlattas, and K. Harding. J . Am. Cheni. SOC.,91, 535 (1969). 39. E.J. Corey and E. Hamanaka. J. Atti. Chein. SOC.,89,2758 (1967). 40. E. J. Corey. In W. 0. Milligan, Ed., Proceedings of the Robert A. Welch Foundation Corferences on Chemical Research. XII. Organic Synthesis, Houston, Texas, 1969. pp. 51-79.
140
The Total Synthesis of Prostaglandins
41. E. J. Corey, Z. Arnold, and J. Hutton. Tetrahedron Lett., 1970, 307. 42. E. J. Corey and R. Noyori. Tetrahedron Lett., 1970, 311. 43. E. J. Corey, R. Noyori, and T. K. Schaaf. J. Am. Chem. SOC.,92,2586 (1970). 44. E. J. Corey. T. K.Schaaf, W. Huber, U. Koelliker, and N. M. Weinshenker.J. Am. Chetn. Soc., 92, 397 (1970). 45. E. J. Corey. Ann. N . Y . Acad. Sci.,180, 24 (1971). 45a. J. A. Dale, D. L. Dull, and H. S. Mosher. J . Org. Chem., 34, 2543 (1969). 46. E. G. Daniels, J. W. Hinman, B. A . Johnson, F.P. Kupiecki, J. W. Nelson, and J. E. Pike. Biochem. Biophys. Res. Coinmrm.,21, 413 (1965). 47. E. G. Daniels, W. C. Krueger, F. P. Kupiecki, J. E. Pike, and W. P. Schneider. J. Am. Chem. SOC., 90, 5894 (1968). 48. E. G. Daniels and J. E. Pike. In P.W. Ramwell and J. E. Shaw, Eds., Prostaglandin Symposium of /he Worcester Foundation for Experitnenral Biology, Interscience, New York, 1968, pp. 379-387. 49. D. A. van Dorp, R. K. Beerthuis, D. H. Nugteren, and H. Vonkeman. Nature, London, 203, 839 (1964). 50. D. A. van Dorp, R. K. Beerthuis, D. H. Nugteren, and H. Vonkeman. Biochim. Biophys. Acra, 90,204 (1964). 51. D. A. van Dorp. Nalurwiss., 56, 124 (1969). 52. D. T. Downing, D. G. Ahern, and M. Bachta. Biochem. Biophys. Res. Commun., 40, 218 (1970). 53. N. Finch and J. J. Fitt. Tetrahedron Lett., 1969,4639. 54. S . H. Ford and J. Fried. L f e Sci., 8 (part l), 983 (1969). 55. J. Fried, S . Heim, P. Sunder-Plassman, S. J. Etheredge, T. S. Santhanakrishnan, and J. Himizu. In P. W. Ramwell and J. E. Shaw, Eds., Prostaglandin Symposium of the Worcester Foundation for Experimental Biology, Interscience, New York, 1968, pp. 351-363. 56. J. Fried, S. Heiiu, S. J. Etheredge, P.Sunder-Plassnian, T. S. Santhanakrishnan, J. Himizu, and C. H. Lin. Chetn. Cominrm., 1968, 634. 57. J. Fried, T. S. Santhanakrishnan, J. Himizu, C. H. Lin, S. H. Ford, B. Rubin, and E. 0. Grigas. Nature, London, 223, 208 (1969). 58. J. Fried, M. M. Mehra, W. Kao, and C. H. Lin. Abstracts, 5th Middle Atlantic Regional Meeting, Am. Chem. SOC.,Newark, Del., 1-3 April 1970, p. 60. 59. J. Fried, M. M. Mehra, W. L. Kao, and C. H. Lin. Tetrahedron Lett., 1970,2695. 60. J. Fried, C. H. Lin, M. M. Mehra, W. L. Kao, and P. Dahren. Ann. N . Y. Acad. Sci., 180, 38 (1971). 61. E. Granstrom and B. Samuelsson. J. Am. Chem. SOC.,91, 3398 (1969). 62. K. GrQn. Chem. Phys. Lipids, 3, 254 (1969). 62a. K. GrQn and B. Samuelsson. J. Lipid Res., 5, 117 (1964). 63. M. Hamberg. Eur. J. Biochem., 6, 147 (1968). 64. M. Hamberg and U. Israelsson. J. Biol. Chem., 245, 5107 (1970). 65. M. Hamberg and B. Samuelsson. J. Am. Chem. SOC.,91,2177 (1969). 66. E. Hardegger, H. P. Schenk, und E. Borger. Helu. Chim. Acta., 50, 2501 (1967). Other work including model studies by this group is found in Dissertations. Nr.
References
141
3794, W. Graf; 3796, H. A. Kindler, 3870, P. Muller; 3942, F. Naf; 3943, E. A. Broger; 4003. H. Schenk; and 4006,J. Vonarburg, E. T. H. Zurich. 67. J. W. Hinman. Postgrad. Med. J., 46, 562 (1970). 67a. K. G. Holden, B. Hwang, K. R. Williams, J. Weinstock, M. Harman, and J. A. Weisbach. Tetrahedron Lett., 1968, 1569. 68. E.W. Horton. Physiol. Rev., 49, 122 (1969). 69. W. Jubiz and J. Frailey. Clin. Res., 19, 127 (1971). 70. G.Just and C. Simonovitch. Tetrahedron Lett., 1967, 2093. 71. G.Just, C.Simonovitch, F. H. Lincoln, W. P. Schneider, U. Axen, G. B. Spero, J. E. Pike. J. Am. Chem. Soc., 91, 5364 (1969). 72. J. Katsube and M. Matsui. Ag. Biol. Chem., 33, 1078 (1969). 73. R. Klok, H.J. J. Pabon, and D. A. van Dorp. Rec. Trau. chim. Pays-Bas Belg., 87, 813 (1968). 74. R. Klok, H. J. J. Pabon, and D. A. van Dorp. Rec. Trau. chim. Pays-Bas Belg., 89, 1043 (1970). 75. 0.Korver. Rec. Trau. chim. Pays-Bas Belg., 88, 1070 (1969). 76. C.Larsson and E. AnggArd. Acta Pharmacol. Tox., 28 (Suppl. l), 61 (1970). 77. L. Levine and H. Van Vunakis. Biochem. Biophys. Res. Conrrnun., 41, 1171 (1970). 78. M.Miyano. Tetrahedron Left., 1969,2771. 79. M . Miyano and C. R. Dorn. TetrahedronLett., 1969, 1615. 80. M. Miyano. J . Org. Chem., 35,2314 (1970). 81. R. B. Morin, D. 0. Spry, K.L. Hauser, and R. A. Mueller. Tetrahedron Lett.., 1968,6023. 82. J. Nakano, E.AnggArd, and B. Samuelsson. Eur. J. Biochem., 11,386 (1969). 83. J. Nakano, E. AnggArd, and B. Samuelsson. Pharmacologist, 11,238 (1969). 84. D. H. Nugteren, R. K. Beerthuis, and D. A. van Dorp. Rec. Trau. chim. Pays-Bas Belg., 85,405 (1966). 85. D.H.Nugteren, D. A. van Dorp, S. Bergstrom, M. Hamberg, and B. Samuelsson. Nature, London, 212, 38 (1966). 86. D. H. Nugteren, H. Vonkeman, and D. A. van Dorp. Rec. Trau. chim. Pays-Bas Belg., 86, 1237 (1967). 87. D. H. Nugteren. Biochim. Biophys. Acta, 210, 171 (1970). 88. D. E. Orr and F. B. Johnson. Can. J . Chem., 47,47 (1969). 89. H. J. J. Pabon, L. van der Wolf, and D. A. van Dorp. Rec. Trau. chim. Pays-Bas Belg., 85, 1251 (1966). 90. C. Pace-Asciak and L. S. Wolfe. Chem. Commun., 1970, 1234. 91. C.Pace-Asciak and L. S . Wolfe. Chem. Commun., 1970,1235. 92. R. Pappo, P. W. Collins, and C. J. Jung. Ann. N . Y.Acad. Sci., 180, 64 (1971). 93. J. E.Pike, F. P. Kupiecki, and J. R. Weeks. In S. Bergstrom and B. Samuelsson, Eds., Nobel Symposium 2, Prostaglandins, Almqvist and Wiksell, Stockholm, 1967,pp. 161-171. 94. J. E. Pike, F. H. Lincoln, and W. P. Schneider. J. Org. Chem., 34. 3552 (1969). 95. P. W. Ramwell, J. E. Shaw, G. B. Clarke, M. F. Grostic, D. G. Kaiser, and J. E.
142 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 106. 107.
108. 109. 110. 111. 112. 113. 114.
115.
116. 117. 118.
119. 120. 121.
The Total Synahesis of Prostaglandins Pike. In R. T. Holman, Ed., Progress in the Cheftiistry of Furs and Orher Lipids, Vol. 9, Pergamon Press, Oxford, 1968,pp. 231-273. P. W. Ramwell, J. E. Shaw, E. J. Corey, and N. Andersen. Nature, London, 221, 1251 (1969). P. W. Ramwell and J. E. Shaw. Recent Prog. Horrn. Res., 26, 139 (1970). B. Samuelsson and G . Stallberg. Acta Chern. Scand.. 17, 810 (1963). 9. Samuelsson. Angew. Cheiti. l n t . Ed., 4,410 (1965). B. Samuelsson. J . Ani. Chetri. Soc., 85, 1878 (1963). B. Samuelsson. Abstract, 4th Int. Congr. Pharmac., Basle, 14-18July 1969,p. 11. B.Samuelsson. Proc. 4th Int. Congr. Pharmac., Basle, 14-18 July 1969,Vol. 4, pp. 12-31,Schwabe, Basle, 1970. B. Samuelsson, M. Hamberg, and C. C. Sweeley. Anal. Biochern., 38, 301 (1970). W. P. Schneider. CIieni. Cornmiin., 1969, 304. W.P. Schneider, U.Axen, F. H. Lincoln, J. E. Pike, and J. L. Thompson. J . Am. Clierrr. SOC.,90, 5895-5896 (1968) [see erratum 91, 1043 (1969)l. W.P. Schneider, U. Axen, F. H. Lincoln, J. E. Pike, and J. L. Thompson. J. Am. Clrenr. Soc., 91, 5372 (1969). W. P. Schneider and J. Muenzer. Unpublished data. Useful differences were seen in both the proton and fluorine magnetic resonance spectra. I t has been suggested by G . Slomp of The Upjohn Company that the relevant peaks in the fluorine spectra would be sharper and more easily integrated if the corresponding tri-deuteromethoxy reagent were used. V. Schwarz. Coll. Czech. Chern. Cortim., 26, 1207 (1961). H.Shio, N.H. Andersen, E. J. Corey, and P. W. Ramwell. Abstracts, 4th Int. Congr. Pharmac. Basle, 14-18 July 1969, p. 100. H. Shio, P. W. Ramwell, N. H. Andersen, and E. J. Corey. Experientia, 26, 355 (I 970). C.J. Sih, G. Ambrus, P. Foss, and C. J. Lai. J. Am. Chettr. Soc., 91, 3685 (1969). C.J. Sih. C.Takeguchi, and P. Foss. J . A m Chern. Soc., 92,6670 (1970). D. P. Strike and H. Smith. Tetrahedroti Lett., 1970,4393. C. 9. Struijk. R. K.Beerthuis, H. J. J. Pabon, and D. A. van Dorp. Rec. Trao. chim Pays-Bas Belg., 85, 1233 (1966). D.Taub, R. D. Hoffsommer, C. H. Kuo, H. L. Slates, Z. S. Zelawski, and N. L. Wendler. Cherri. Cornmion., 1970, 1258. M.Vandewalle, V. Sipido, and H. DeWilde. Bitll. SOC.C h h . Belg., 79, 403 (1970). H.Vonkeman, D.H . Nugteren, and D. A. vanDorp. Biochitti. Biophys. Acra, 187,581 (1 969). A. J. Wcinheimer and R. L. Spraggins. Abstracts, 158th Meeting Am. Chem. SOC., New York, 8-12 September 1969, MEDI 41. A. J. Weinheimer and R . L. Spraggins. Tetrahedron Let,., 1969, 5185. A. J. Weinheimer and R. L. Spraggins. In H. W. Youngken, Jr., Ed., Food-Drugs from rhe Sea, Proc. Marine Technology Society 1969, 1970,pp. 311--318. Y.Yura and .I.Ide. Cheni. Pliarm. B u l l . , Tokyo, 17, 408 (1969).
Total Synthesis OfNatural Products, Volume1 Edited by John Apsimon Copyright © 1973 by John Wiley & Sons, Inc.
The Total Synthesis of Pyrrole Pigments A. H. JACKSON Department of Chemistry, University College, Cardiff
AND
K. M. SMITH Robert Robinson Laboratories, Uniuersity of Liverpool
Introduction Nomenclature Strategy of Porphyrin Synthesis Pyrroles A. Formation of C-N Bonds Only B. Formation of 3 4 and C-N Bonds C. Formation of 2-3 and C-N Bonds D. Formation of 2-3 and 4-5 Bonds E. Pyrroles from Other Heterocycles F. Modification of Pyrroles for Use as Intermediates 5. Dipyrrolic Compounds A. Pyrromethanes (Dipyrrylmethanes) B. Pyrromethenes (Dipyrrylmethenes) C. Pyrroketones (2,2’-Dipyrrylketones) 6. Porphyrins from Mono- and Dipyrrolic Precursors A. From Pyrroles B. From Pyrromethanes C. From Pyrromethenes D. From Pyrroketones
1. 2. 3. 4.
144 144 148 149 150 150 152 153 154 154 161 161 163 164 166 166 167 169 176
143
The Total Synthesis of Pyrrole Pigments
144
Porphyrin Synthesis from Open-Chain Tetrapyrrolic Intermediates From a,c-Biladienes B. From n-Oxobilanes C. From b-Oxobilanes and Oxophlorins D. From b-Bilenes 8. Chlorins and Other Partially Reduced Porphyrins 9. Bile Pigments 10. Prodigiosin and Related Compounds 1 1. Corrins and Vitamin B,, A. Eschenmoser's Approach B. The Woodward-Eschennioser Approach C. Cornforth's Approach 12. Postscript References
7.
A.
1.
119
179 184 187 194 199 209 221 232 234 245 26 1 269 271
INTRODUCTION
In this essay we review the main methods currently available for the synthesis of naturally occurring porphyrins and bile pigments, as well as the structurally and biosynthetically related vitamin B,, (cobalamin). A short section is devoted to the small group of tripyrrolic bacterial pigments of which prodigiosin is the characteristic example. Heme and chlorophylls u and b are the most widespread natural pigments and perform a complementary role in nature, being associated with the oxidative and energy-liberating processes of plant and animal metabolism on the one hand and the reduction and energy-trapping processes of photosynthesis on the other. As well as being the oxygen carrier of hemoglobin, heme is also the prosthetic group in many of the cytochromes, catalases, and peroxidases, and modified hemes occur in the prosthetic group of cytochromes a, and u3, and c . Other important naturally occurring porphyrins include the algal chlorophylls c , d, and e, the bacterial chlorophylls such as bacteriochlorophyll, and the Chlorobiirni chlorophylls (650) and (660). With the notable exceptions of heme and chlorophylls a and b most of these pigments have not yet been synthesized and their structures rest upon comparisons of their degradation products, with other porphyrins of importance in relation to heme and chlorophyll biosynthesis. Some of the more important earlier reviews in this area are given in references 1-8,
2. NOMENCLATURE
Two systems are currently in use for numbering the porphyrin nucleus and related open-chain polypyrroles, and their relative merits are now under
2. Nomenclature
145
active discussion with a view to choosing a unified convention. The earlier system2 devised by Hans Fischer is shown in 1 and the new system devised ~ include vitamin B,, and by the IUPAC Nomenclature C ~ m m i t t e e ,to related macrocycles in the same scheme as porphyrins, is shown in 2.
20
18
Y
7.
1
0
17
16
13
2
The Fischer system is used in this review for porphyrins, to provide continuity with the massive chemical and biological literature of the past and because it is still used by the majority of chemists and biochemists working in the field today. However, the IUPAC system (2) is used for vitamin B,, and other “corrinoid” compounds. The chlorophylls present in most green plants and the Chlorobium chlorophylls are dihydroporphyrins or chlorins, in which one of the peripheral double bonds has been reduced; c~nventionally,~ their structures have always been written with the D ring partially reduced (3). Bacteriochlorophyll, a tetrahydroporphyrin, is usually written with the peripheral linkages in rings B and D reduced. The recently discovered dihydroporphyrin system (4), in which hydrogen is added to one of the meso positions and to nitrogen, has been named phlorin.1°
3
4
The nomenclature for di-, tri-, and tetrapyrrolic compounds in which the pyrrole rings are linked together by methane (-CH,-) (Sa), methene (5b), or carbonyl (-CO-) (5c) bridges has not yet been fully (-CH=) systematized. The accepted numbering systemlo2 for the dipyrrolic compounds
146
The Total Synthesis of Pyrrole Pigments
is shown in 5, and these compounds should strictly be referred to as 2,2’-
5a
5b
5c
dipyrrolyl methanes, methenes, or ketones. These names frequently are abbreviated to dipyrromethane, dipyrromethene, and dipyrroketone, or to the even simpler pyrrometlrane, pyrromethene, and pyrroketone; the latter convention is used in this review. (This does not normally cause any confusion because few of the 2,3’- and 3,3’-analogs are known, and moreover they are not known to have any biological significance.) The numbering system2shown in 6 is used for the tetrapyrrolic compounds, although some authors6 have used an alternative system based on the IUPAC nomenclature for vitamin B12 and porphyrins. The generic name bilane was originally applied2 to the tetrapyrrolic bile pigments (6) with three methylene (CH,) bridges. Subsequently, Lemberg“ used this name for the corresponding 1’,8‘-dioxygenated tetrapyrrolic bile pigments, but many synthetic non-
6
oxygenated analogs are now known and to refer to these as “dideoxybilanes” seems rather cumbersome. For this reason the name bilane is used generically in this review for all the compounds with the carbon nitrogen skeleton shown in 6,and in accord with Lemberg” the related tetrapyrroles with one or more methene linkages are referred to as bilenes, biladienes, and bilatrienes. These compounds are often referred to in the literature as ‘‘linear’’ tetrapyrroles, but the term “open-chain” tetrapyrroles is to be preferred, because space filling models clearly show that the Psubstituents in the individual pyrrole nuclei tend to force the adoption of a coiled conformation. For this reason also the structures are written as in 5 and 6 to emphasize this point. As with other natural products, a wide variety of trivial names are in current use for both porphyrins and bile pigment^.^^^ A further complication arises because of the possibilities for isomerism about such a symmetrical nucleus
2. Nomenclature
147
as that in porphyrins, and Fischer2 devised a system for numbering isomeric compounds (Table l), especially the simpler degradation products of the Table 1. Trivial Names of Porphyrins Related to Heme and Chlorophyll
Substituent Combinations in the /%Positions 1-8
Trivial Name
4( Me, Et)
Etio-porphyrins Copro-porphyrins Uro-porphyrins Meso-porphyrins Proto-porphyrins Deirtero-porphyrins Pyrro-porphyrins Rhodo-porph yrins A = CH,CO,H,
Number of Isomers 4 4 4
4 w , P) 4(A, P) 2(Me, Et), 2(Me, P) XMe, V, XMe, P) W e , W ,2(Me, P) ](Me, H), 2(Me, Et), l(Me, P) l(Me, C02H), 2(Me, Et), 1(Me, P)
P
=
CH,CH&02H,
V
=
15 15 15
25 25
CH-CH,
natural materials. There are four possible etioporphyrins, for example, in which each pyrrole ring bears one methyl and one ethyl substituent, and the isomer designated as etioporphyrin-111 can not only be obtained by reduction of heme but is also closely related to the chlorophylls. Nearly all naturally
Me
M
Et
e NH a N E
N
Me Et
HN
Et
Et
Et
Et
NH
t
N
Me Et
Me
Etioporphyrin type isomers.
N
HN
Me
148
The Told Synthesis of Pyrrole Pigments
occurring porphyrins and the corrin chromophore of vitamin B,, can be related, at least formally, to etioporphyrin-111 and are often referred to as type 111 compounds. Small amounts of type I porphyrins occur in certain rare pathological conditions, but type I1 and IV porphyrins have not been found in nature. There are also four isomers of the copro- and uro-porphyrins, but when there are three different substituents as in the related meso-, proto-, and dezcfero-porphyrins, then 15 isomers are possible; with the pyrro- and rhodoporphyrins which have four different substituents, the number of positional isomers rises to 25 (Table 1). The substitution pattern in the type 111 isomers of meso-, proto-, deutero-, yyrro-, and rhodo-porphyrins is indicated schematically in 7.
Mea R
Me
I
I
R = R’ = P R = Et; R‘ = P prom-1X R = V ; R‘ = P derrtero-lX R = H ; R‘ = P pyrro-XV R = Et; R’ = H rliodo-XV R = Et; R‘ = CO,H copro-It1 rncso-lX
Me
Me
P
1
3.
R‘
STRATEGY OF PORPHYRIN SYNTHESIS
In reviewing recent progress5,’,*made in the synthesis of porphyrins it seems appropriate to consider the general philosophy and strategy underlying the various methods now available and to cite individual naturally occurring porphyrins as examples. The most widely used porphyrin synthesis to date has been the fusion of two pyrromethene units ;2 Fischer’s classical synthesis of heme is perhaps the most well-known cxample. However, in recent years the pre-eminence of this method has been challenged by the development of new “stepwise” methods for the rational synthesis of open-chain (or “linear”) tetrapyrroles, which can then be cyclized to porphyrins. This approach has proved increasingly attractive since Corwin and Coolidge’s original preparation and cyclization of an open-chain bilane to etioporphyrin-I1 nearly 20 years ago,I2 although recent work has cast some doubts on the isomeric purity of this product (see below). Mild methods are now available for the preparation of bilenes,l3 biladienes,I4 and a- and 6-oxobilanes,l6 all of which
4. Pyrroles
149
can be cyclized to porphyrins under mild conditions. None of these methods is truly stepwise, for they all involve the coupling of two dipyrrolic units, but there do not seem to be any advantages in synthesizing the tripyrrolic compounds as intermediates;la indeed, there may be positive disadvantages if the tri- or tetrapyrrolic compounds are relatively unstable. One other attraction of these new stepwise methods is that they generally involve much milder reaction conditions than the Fischer pyrromethene synthesis,2and consequently some of the more labile side chains in the desired porphyrin can be carried through from the pyrrole stage, either directly or in a modified or protected form. This is generally more efficient, but as will be seen it is still occasionally necessary to introduce a substituent at a late stage in this synthesis after macrocycle formation, and the position of insertion may rcquirc blocking during the mono-, di, and tetrapyrrolic stages, with, for example, bromine or iodine, which can subsequently be removed by hydrogenolysis. The synthesis of chlorins (dihydroporphyrins) and tetrahydroporphyrins such as the chlorophylls and the bacteriochlorophylls presents additional problems. Extra hydrogen atoms can be introduced by reduction of porphyrins in various ways, but the stereochemistry and position of addition are difficult to control. Since pyrroles are the basic building blocks for all the present porphyrin syntheses we first discuss briefly the synthesis of monopyrroles and their coupling to give the various dipyrrolic intermediates, pyrromethanes, pyrromethenes, and pyrroketones. The direct synthesis of porphyrins from the mono- and dipyrrolic compounds is also discussed, followed by the coupling of dipyrrolic compounds to form tetrapyrrolic intermediates and their conversion into porphyrins.
4. PYRROLES
In choosing a synthetic approach to a given pyrrole, symmetry factors within the molecule are of primary importance, and a choice is made from one of the general routes1' outlined in this section. The large majority of the methods discussed are from noncyclic precursors, and so the synthetic plan is variable with respect to the points at which the pyrrole ring (8) may be closed.
8
150
The Total Synthesis of Pyrrole Pigments
A.
Formation of C-N
Bonds Only
Typically, this method involves the much used reaction of a primary amine, or ammonia, with a 2,Sdiketone:
Known in general terms as the Paal-Knorr method, this synthesis can give very good yields of pyrroles and is limited mainly by the inaccessibility of suitable diketones. It has not found a great deal of application i n the preparation of pyrroles for contemporary porphyrin syntheskh An example of this general approach is the original, and still useful, synthesis of pyrrole (8) by the pyrolysis of ammonium mucate (9) and pyrrole itself is still much
H
HOCH-CHOH
H
a
9
used i n porphyrin synthesis because when it is polymerized in the presence of aldehydes (the Rothemund18 reaction) it furnishes the corresponding ?)ieso-O:,,!I ,y ,&tetrasubst i t u ted porphyri n.
B. Formation of 3 4 and C-N
Bonds
This approach includes the classical (and without doubt most frequently used) avenue to pyrroles, the Knorr synthesis. In its most popular form this method gives an excellent route to 2,4-dialkylpyrrole-3,5-dicarboxylates[e.g., “Knorr’s pyrrole” (14)]. Thus condensation of an a-aminoketone (12) with a ketone (13) possessing an active methylene group (usually a p-ketoester) in acetic acid buffered with sodium acetate furnishes pyrroles (e.g., 14) in good yield. The u-aminoketones (e.g., 12) are prepared in siru by zinc (or sodium dithionite) reduction of the corresponding oximinone 11 prepared earlier from the appropriate ketone [or 0-ketoester (lo)] with nitrous acid. The synthesis is o f wide application, which accounts for its popularity through the years. The comparatively recent introduction of f-butyl and benzyl esters as protecting groups has increased the versatility of this synthesis even further,
Me
\6
-
0
C
CH,
EtOzC
-
Me C
I
CH
/ \
N H,
12
J
-
0
C I
Zn
d
N
\
OH
11
/cozEt
CH, IC
+
/ \
0
151
HOAc
/ \
Et0,C
10
Et0,C
\ /
HONO
I
/
Me
4. Pyrroles
g:
Me Et0,C
___f
H
Me
14
13
since differential protection of the two carboxylates expedites the specific manipulation of these functions at later stages. The 5-substituent can be hydrogen, but yields are better when it is an acyl or carboxylate group; yields are poor when the 3-substituent is alkyl. The most important limitation to Knorr’s synthesis is the inherent symmetry of the products (e.g., 14). The Hantzsch synthesis involves the condensation of a p-ketoester (15) and an a-chloroketone (17) or a-chloroacetaldehyde in the presence of ammonia; it seems likely that 2-aminocrotonate (16) is an intermediate in EtOzC
EtOzC
\
\
‘HZ
I
C
/ \
Me
15
0
NH,
CH
+
II
C
/ \
Me
16
CHZ
NH,
I
C
/ \
0
17
- aN8t,, EtO,C Me
Me
H
this reaction. It has long been thought to be far less generally applicable than the Knorr synthesis, but recently the Hantzsch synthesis has been extended by MacDonald,l9 who showed that a-halo derivatives of aldehydes other than acetaldehyde may be used, and also that in certain cases benzyl and tbutyl pyrrole-4-carboxylates are accessible. Hitherto, only eight or nine different pyrroles had been made by the Hantzsch method, but the extension to the use of aldehydes (which are much more easily halogenated specifically than are ketones) has increased the general applicability of this route.
152
The Total Synthesis of Pyrrole Pigments
C. Formation of 2-3
Bonds
and C-N
In attempting to modify the Knorr synthesis, Fischer and Finkz0discovered that condensation of @-ketoaldehydes (in the form of their acetals (19)] and the usual Knorr oximinones (e.g., 18) in the presence of zinc and buffered acetic acid gave acceptable yields (-40%) of pyrroles of type 20. An acetyl group is lost in the process, showing that, rather unexpectedly, the 2-3 and Me0
OMe
I
OH 18
20
19
C-N bonds had been formed. KleinspehrP found that this method could be modified by use of oximinomalonic ester or 2-oximino-l,3-diketones rather than the oximinoketoesters (18). The route was then modified even further by Johnson,22 who recommended the use of oximinoacetoacetic esters (21) (particularly the benzyl and r-butyl esters) or a-oximino ketones in condensation with 1,3-diketones (22). This modification is probably one of the most Me
\
0 MeCO
+
',
-.k C
/ C--CH
I
C
R'
dMe
Me Zn/HOAc
-
.I\ 0 Me
RO,C
H
22
OH
21
frequently used routes to pyrroles in modern synthesis, due to the easy accessibility of suitable 2-substituted 1,3-diketones (22). A further method developed by Fischer and Finkz0 utilized the hydroxymethylenernethyl ethyl ketone 23 and oximinoacetic ester 18, furnishing the pyrrole 24 after the usual zinc reduction.
4. Pyrroles
Na+ 0-
\
+
MeCO
\
EtO&
CH-CH
c - +
0
EtO&
Me
H
N
I OH
18
H
I
/ \
C
/ \
/Me
153
23
24
Kenner and co-workers have reported a pyrrole synthesis utilizing N-tosyl glycine derivative^.^^ Thus base-catalyzed addition of N-tosyl glycine (25) to R'
Me
>-<
0
CHZ
26
NH I
/
Tos
CH,
___+
\
R17&:2R I
Tos
COzR
27
25
X:02R - kC:02R R'
H
I
H
Tos
28
29
R1 = H or Me R = Et, But, or CH,Ph
an a,,!?-unsaturated ketone (26) gave 27, which, after dehydration to the A3pyrroline 28, afforded the corresponding pyrrole-2-carboxylate 29.
D. Formation of 2-3
and 4-5
Bonds
Very few useful examples of this general method exist. Perhaps the best is the preparation of N-substituted pyrroles (32) by condensation of a suitable
154
The Total Synthesis of Pyrrole Pigments
tri-substituted amine (30) with an a-diketone (31). As might be expected, the
R3 \- / / / \
R2
0 R'-CH,
31
0
\ / CH,-R4 N R
- XR4 R'
R
32
30
pyrroles obtained directly from this method, because of their N-substitution. are of little use as intermediates in porphyrin synthesis.
E. Pyrroles from Other Heterocycles The best known method24of this type is the treatment of furans (e.g., 33) with ammonia or substituted amines, at temperatures between 300 and 500". The moderate yields, limitations due to availability of furans, and the drastic
XMeRNIi?
Me
-330'
Me
R
Me
33
conditions all combine to make this method of limited value for preparation of the diverse pyrroles required for porphyrin synthesis. F. Modification of Pyrroles for Use as Intermediates Even with the large number of general methods for pyrrole synthesis, which have been outlined briefly, there are still many cases in which a required pyrrole cannot be obtained directly by ring synthesis. In such cases, pyrroles obtained from the Knorr synthesis or one of its modifications are usually used as substrates for further elaboration. "Reductive C-alkylation" is a recently reported methodz5 by which pyrroles from existing syntheses can be modified to give access to pyrroles which, due to deficiencies in these very syntheses, were not readily available. When a pyrrole is treated with hydriodic acid and paraformaldehyde at 100" it is C-methylated, alkoxycarbonyl and acetyl groups being lost. In
4. Pyrroles
155
hydriodic acid at lower temperatures (1 5-45’) typical pyrroles retain their labile groups and all free positions are C-alkylated (methylated if formaldehydc is used or else otherwise alkylated depending on the carbonyl agent). Apart from providing a route to diverse alkylated pyrroles, it seems that this method may also be applicable to the degradative structure determination of porphyrins.2e The “rational” synthesis of all four uroporphyrin isomers was not achieved until recently by M a c D ~ n a I d , ~ ’due - ~ ~to diffi~ulties~~ involved in the preparation of the required pyrroles (e.g., 34) rather than thc construction of the
34
Et0,C 0
Et02C
B
I
CH2-C
C
Et0,C
/ \
CH2
+
I
N
I
OH
I
I
C
O
/ \
Me
kLc
CH,
/CozCHzPh
+
Et 02C
H
35a R = CO,CH,Ph 35b R = C O , H 3% R = H 35d R = CHO 35e R = CH=CHC02H
porphyrin macrocycle itself. The problem was overcome by first synthesizing the pyrrole 35a from benzyl acetoacetate and oximinoacetone dicarboxylic ester (a typical example of the Knorr method). Catalytic hydrogenation gave the p-carboxylic acid 35b, which was thermally decarboxylated to the corresponding &free pyrrole 35c. Formylation gave 356, which was elaborated to the acrylic pyrrole 35e by means of a Knoevenagel condensation with malonic acid. Catalytic hydrogenation and esterification gave the pyrrole 34, an invaluable intermediate in the synthesis of the uroporphyrin isomers. Without doubt, the most important and synthetically desirable pyrrole in recent times has been porphobilinogen (PEG) (36).Since its isolation32from the urine of patients suffering with acute porphyria, it has been found to be in the biogenesis of the blood and plant pigments as well a direct as the biliproteinoids and vitamin BIZ.Any truly useful synthesis of PBG
The Total Synthesis of Pyrrole Pigments
156
HOSC
CH,CO,H
I
I
NH2CH,
€I 36
must be capable of modification to allow introduction of isotopic labels at specific sites within the molecule, for exploitation in biosynthetic investigations. Already, much work has been done with PBG obtained enzymatically from labeled 6-aminolevulinic acid, but further investigations, particularly the examination of the pathway by which PBG is polymerized to the unsymmetrical type I11 uroporphyrinogen, require more refined labeling at given
H0,C
CH,CO,Et
I
CH,CO,H
I
I
CO,H
H
37a R = CH, 37b R = CHO 37c R=CH=NOH
38
37d R =CH,NH,
H
39a R = CO,H 39b R = H
HO2C
I
CH,CO,H
I
I
I
HC
OHC R = CO,H 40b R = I
C€€,CO,H
HOiC
II
N
40s
HO’ 41
4. Pyrroles
157
positions, and such substrates can be obtained only by total synthesis. In a series of papers, MacDonald and co-workersg4described the synthesis of PBG, with gradually increasing efficiency, based on a rather classical overall plan of attack. Thus the pyrrole 37a was dichlorinated with sulfuryl chloride and the 2-formylpyrrole 37b obtained by hydrolysis. This was converted to the oxime 37c by standard methods from which the aminomethylpyrrole 37d was accessible by catalytic hydrogenation. Hydrolysis gave the tricarboxylic acid 38, which was decarboxylated most readily as the lactam 39s to PBG lactam (39b). Alkaline hydrolysis gave PBG (36) in an overall yield of 5 % from the pyrrole 37a. This was later improved by hydrolysis of the formyl triester 37b to the corresponding tricarboxylic acid 40a followed by iodinative decarboxylation to 40b; treatment with hydroxylamine afforded the pyrrole oxime 41 with concomitant loss of the iodine. Alternatively the formyl acid (40a), could be converted directly into the oxime (41) by treatment with hydroxylamine. Catalytic hydrogenation of 41 gave porphobilinogen (36) in 29% overall yield from the pyrrole 37a. Plieninger, Hess, and Ruppert3' recently improved the early stages of the MacDonald synthesis of porphobilinogen by a more efficient synthesis of 37a.
x;02 CHtCOZEt
Et02C
I
CHzC02Et
,
Mc
I
H
37a
43
A0
Me
+
N/c(Co2Et)2
I OH
42
44
Their condensation of diethyl oximinomalonate (42) with the 2,4-diketone 43 in presence of zinc and sodium acetate gave 37a in 41 %yield, with none of the unwanted isomer (44) being detected. A most ingeniously conceived synthesis of PBG based on its readily reversible conversion to the lactam 39b has recently been reported by Rapoport and ~o-workers.~'' The starting material, a pyridine (45), was converted to PBG
158
The Total Synthesis of Pyrrole Pigments
(36) in an overall yield of 19 %, a considerable improvement on existing routes. In addition, the totally different concept in the synthetic approach outlined below provides entirely different opportunities for incorporation of radioactive tracers; the high overall yield also makes it possible to contemplate the incorporation of tracers at unusually early stages. Thus the trisubstituted pyridine 45 was chosen as the starting material, the nitrogen atom of the nitro C0,Et
I
4s
46
“‘“m cop
47
H ,CH ,CO,1-¶
H 48
49
CO,H CH,CN
I
M e o a s c O * H
ti 50
51
group being destined to become the nitrogen atom of the pyrrole ring in PBG. After suitable modification, the pyridine ring itself was to become the lactam ring in PBG lactam (39b), with the pyridine 4-methyl group and its acidic hydrogens being the site for initial connection of the future pyrrole ring. Treatment of 45 with diethyl oxalate and sodium ethoxide gave 46, which cyclized to the azaindole 47 on reduction of the nitro group. The 1-position of indoles is well known for its vulnerability to electrophiles, and so the required propionate side chain was constructed by means of a Mannich reaction (to give 48) followed by treatment with diethyl sodiomalonate, furnishing 49 after exhaustive hydrolysis. Catalytic reduction of the newly created pyridone ring gave the pyrrole 39a, an intermediate in MacDonald’s
4. Pyrroles
159
approach, which could be decarboxylated to 39b and then hydrolyzed to PBG. This general approach has been extended to the synthesis of PBG analogs such as 51; R = H and 51; R = CH2COOH by direct transformation of the azaindoles, 47 and 50, respectively; the latter was accessible by treatment of the quarternary salt of the Mannich base 48 with sodium or potassium cyanide. Without exception, the methods for producing polypyrroles utilize ionic reactions, and the pyrrole ring is the nucleophile. Because of the “rr-excessive” nature of ~ y r r o l ethe , ~ most ~ common nucleophilic species are 2-unsubstituted pyrroles (52). In certain cases, and particularly when 3-, 4-, and 5-substituents are not electron-withdrawing functionalities, the corresponding 2carboxylic acid 53 may be used instead. Such pyrroles (53) are readily accessible from the corresponding esters by hydrolysis or hydrogenolysis. It should
XH &
COZH
H
H
52
53
be noted, however, that t-butyl esters of pyrroles, when deesterified with, for example, trifluoroacetic acid, suffer concomitant decarboxylation to 52 and, indeed, 2-unsubstituted pyrroles are often accessible from the corresponding carboxylic acid by either acid-catalyzed or thermal decarboxylation. Pyrroles (e.g., 54) bearing electron-withdrawing groups usually cannot be decarboxylated with such ease and may require “iodinative decarboxylation” involving iodination to 55 followed by hydrogenolysis of the halogen atom over Adams
54
55
catalyst.38 Pyrroles such as 54 are available from 5-methylpyrrole-2-carboxylates (56), (the most common type of pyrrole produced from the classical syntheses,) by trichlorination to 57 with sulfuryl chloride, followed by hydrolysis (see Scheme 1). Mono-halogenation of pyrroles (56) furnishes the a-halomethylpyrroles 58, a useful electrophilic species in pyrromethane
160
The Total Synthesis of Pyrrole Pigments
synthesis. Further chlorination to the dichloromethylpyrroles 59 followed by hydrolysis furnishes the corresponding a-formylpyrroles 60, which are used extensively in pyrromethene synthesis. Alternatively, acetoxylation of 56 with lead tetraacetate furnishes the mono- and diacetoxyrnethyl analogs of 58
58
56
1
1 ca. SO,CI,
RO,C &
L o
H
aq. NaOAc
R 0,C a
4 -
60
c , , , 11
I
59 I ea. SO,CI,
aa. NaOAc
RO,C a
C 0 2 H '-
RO,C
H
54
\
SOCI,
RO*C
-4
H
1-1 57
ace,, ROH
HNMcP
RO,C
H
62
61
63
Scheme 1
and 59; the latter can also be hydrolyzed to the formylpyrrole 60. NNDimethyl carboxamidopyrroles (61) have found much application in the synthesis of pyrroketones and are 0btained3~either by treatment of a-trichloromethylpyrroles (57) with dimethylamine followed by hydrolysis or else by
5. Dipyrrolic Compounds
161
treatment of pyrrole a-acid chlorides (62) [obtained from the corresponding a-carboxylic acid (54) with thionyl chloride] with dimethylamine. If the acid chloride 62 is treated with an alcohol R’OH in presence of a base (e.g., N,N-dimethylaniline), the corresponding 2,5dicarboxylate (63) results. 5. DIPYRROLIC COMPOUNDS
A.
Pyrromethanes (Dipyrrylmethanes)
These compounds (Sa), until fairly recent times, were thought to be unstable and not suitable as intermediates in porphyrin synthesis for a variety of reasons. However, a highly successful route to porphyrins devised by MacDonaldle has shown that the supposed limitations of pyrromethanes are not as serious as had been suggested, for example, by Fischer’s almost total reliance on pyrromethenes.2 Symmetrically substituted pyrromethanes (64) can be synthesized in several ways. Treatment of a 2-unsubstituted pyrrole (65) with formaldehyde furnishes pyrromethanes (64) in good yield. The same product can be obtained in even higher yield by refluxing the corresponding 2-bromomethylpyrrole 66a in alcoholic solvents for short periods; a recent modification, due to
NH HN 64
65
H 67
H
J$ H
XCH,
66a X = Br 66b X = O A c
NH H+N
66a X = Br 66b X = OAc
Russian workers,4O utilizes the heating of 2-acetoxymethylpyrroles (66b) in hydrochloric acid/methanol and can furnish the symmetrical pyrromethane 64 in yields as high as 90%. The mechanism of this type of self-condensation presumably involves initial formation of the carbonium ion 67, which is
162
The Total Synthesis of Pyrrole Pigments
attacked by a further molecule of the pyrrole 66. Elimination of the one carbon fragment (probably as formaldehyde) then furnishes the required product. Unsymmetrically substituted pyrromethanes (e.g., 70) are frequently required in porphyrin synthesis, and compounds of this type pose a considerably more complex problem than their symmetrical counterparts. An early solution4' to the problem employed the treatment of a pyrrole Grignard reagent (68) with a pyrrole (69) bearing a potential carbonium ion. Pyrrole
68
X
69
70
= CI, Br, OMe, OAc
65
Grignard reagents are easily obtained by treatment of the corresponding pyrrole 65 with ethyl magnesium bromide. Perhaps the most successful avenue to unsymmetrical pyrromethanes is that developed by MacDonald and c o - w o r k e r ~ .Thus ~ ~ brief heating of 2-unsubstituted pyrroles (65) with 2-substituted pyrroles (69) in refluxingglacial acetic acid buffered with sodium acetate gives good yields of the corresponding unsymmetrically substituted pyrromethanes (70). When the bromomethylpyrrole (66a) is used as theelectrophilic species, it seems likely that this is first converted, in the buffered acetic acid, to the corresponding 2-acetoxymethylpyrrole (66b),quantities of which are often recovered from the reaction. Even when the 2-unsubstituted pyrrole 65 bears an electron-withdrawing function (e.g., CO,R), it is still sufficiently nucleophilic to react to give pyrromethane. The reaction conditions are not so drastic as to cleave I-butyl esters, which gives an added degree of selectivity. In an alternative approach to unsymmetrical pyrromethanes (70), the pyridinium salt 71 of a bromomethyl pyrrole (66a) and the lithium salt 72
5. Dipyrrolic Compounds
163
of a pyrrole a-carboxylic acid are heated in polar solvents (such as methanol/ water or formamide) and good yields of the appropriate unsymmetrically
71
72
substituted pyrromethane (70) are obtained.15sd2A major limitation to this method is the low nucleophilicity of pyrroles (72) having strongly electronwithdrawing substituents. Borohydride reduction of pyrromethenes has found some application, but this is limited by reactions of some substituents with sodium borohydride, and also by the range of synthetically available pyrromethenes.
B. Pyrromethenes (Dipyrrylmethenes) Pyrromethenes are highly colored compounds (free base 73, A,,IRS 450 nm; salt 74, A,,lRx, 500nm) which have found much use in the synthesis of
73
+ 74
porphyrins.2 They are particularly difficult to purify, probably because of their facile protonation and deprotonation; chromatography rarely produces any improvement in purity, and it is therefore a prerequisite that any synthesis must be highly efficient to enable convenient isolation of the pyrromethene. The classical, and still most favored method for pyrromethene synthesis is that due to Fischer,2 i.e. acid-catalyzed condensation of a 2unsubstituted pyrrole (65) with a 2-formylpyrrole (75), which gives excellent yields of the corresponding unsymmetrically substituted pyrromethene salt (74), and phosphoryl chloride has been shown to be a satisfactory reagent for this condensation in certain cases.43 In some cases, the initially formed pyrromethene (74) may undergo further reaction with the 2-unsubstituted pyrrole to form a tripyrrylmethane (76), disproportionation of which may occur in a different way to give a symmetrical pyrromethene (64) rather than the anticipated unsymmetrical product (74). However, this rarely occurs, and the high degree of success with this general method has deterred further developments to find alternative
164
X H
H
H
15
65
*
+,,,x
The Total Synthesis of Pyrrole Pigments
64
x-
+
-6s/1+6s 74
76
routes to pyrromethenes. The treatment of a-unsubstituted pyrroles with formic acid and hydrobromic acid has found application, but this only provides a route to symmetrical pyrromethenes. The controlled oxidation (e.g., with bromine) of pyrromethanes provides a further, but limited, route to pyrromethenes.
C. Pyrroketones (2,2'-Dipyrrylketones) These compounds have found much use in the synthesis of porphyrins and o x o p h l o r i n ~They . ~ ~ are bisvinylogs of amides and do not show any normal
77
65
&MIX H 68
or
&H H 65
+
-+
-$=A+NH HN 79
5.
Dipyrrolic Compounds
165
ketonic behavior (e.g., toward phenylhydrazine or toward reduction with borohydride). They can, however, be reduced3eto the corresponding pyrromethanes with diborane; this is an important reaction in their adaptation as intermediates in porphyrin synthesis. Symmetrical pyrroketones are available by treatment of an a-unsubstituted pyrrole (65) with phosgene, whereas unsymmetrical ones are available from the same nucleophile (65) or its Grignard derivative (68) with a 2-chlorocarbonyl pyrrole (78). A further method of limited applicability is the oxidation of pyrromethanes with lead dioxide and lead tetraa~etate.4~ A reexamination of this method has been reported and it has been found that the same transformation can be brought about more efficiently using bromine followed by sulfuryl chloride, or in some cases by sulfuryl chloride alone.4s A very satisfactory general method for the preparation of pyrroketones based on the Vilsmeier-Haack procedure has been reported.3eThus treatment of 2-N,N-dimethylamidopyrroles (80) with phosphoryl chloride furnishes the
I
80
OPOCl, CI 81
+
C I '
H,O/OAc-
N1-l HN 79
NH HN 83
corresponding complexes (81), which are strong electrophiles and react with 2-unsubstituted pyrroles (82) to give imine salts (83). Hydrolysis gives pyrroketones (79) in overall yields often in excess of 80%.
166
The Total Synthesis of Pyrrole Pigments
6. PORPHYRINS FROM MONO- AND DlPYRROLlC PRECURSORS
A.
From Pyrroles
Polymerization of 2,5-unsubstituted pyrroles (84) with formic acid,46 aldehydes,18 etc., or self-condensation of suitable substituted pyrroles (85), followed by aerial oxidation often affords porphyrins in quite good yields,22*47-56 but unless the 8-substituents in the pyrroles 84 or 85 are identical, a “random” mixture of the four possible porphyrin. isomers is always ~ b t a i n e d . ~ These ~ * ~ ~methods ,~* are therefore clearly of very limited usefulness for the preparation of the unsymmetrical naturally occurring porphyrins, although partial separation of isomeric mixtures such as the coproporphyrins has been achieved by chromatographic method^.^^.^^ However, the mechanism of self-condensation of pyrroles of type 85 is of great interest in relation to the enzymically controlled polymerization of
84
85
X = CI, Br, OAc, NH,, etc.
A
P NII IIN
CH,NH2 36
--+
NI-i H N
A
Uroporphyrin-I11
P
P 86
Uroporphyrinogen-111 A = CHzCO,H
P = CH,CH,CO,H
porphobilinogen (36), which normally leads to uroporphyrinogen-111 (86) (although the centrosymmetrical uroporphyrinogen-I is occasionally observed in rare pathological conditions). Many hypotheses have been advanced to account for the almost ubiquitous occurrence of type IIJ porphyrins in nature
6. Porphyrins from Mono- and Dipyrrolic Precursors
167
(of which uroporphyrinogen-I11 is the precursor), but no experimental evidence to show how one ring is reversed relative to the other three has yet been p r ~ v i d e d . ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ - ~ ~
B. From Pyrromethanes
One of the earliest reported porphyrin synthesese5 from pyrromethanes involved the condensation of the pyrromethane 87a with formic acid in presence of air to give etioporphyrin-11 (88a); more recent applications
, R* , R
NH HN
R R
87
R
= H or CO,H
a b c
R 1 = Me; R2 = Et R'= Me; R2 = P R'= A ; R a = P
88
include the synthesis of copro- and uroporphyrins-I1 (88b and 88c).29-6a However, mixtures of porphyrins are sometimes obtained because under the acidic conditions of the reaction, cleavage and recondensation reaction^^^*^^ may occur at the methane bridges, presumably by the mechanism outlined in Scheme 2. A milder modification developed in Corwin's laboratorya*was the
etc.
+H2
Scheme 2
H+
168
The Total Synthesis of Pyrrole Pigments
use of the N-methyl pyrrole aldehyde 89 instead of formic acid to provide the bridge carbon atoms presumably via tripyrrylmethane-type intermediates (90). However, both procedures arc clearly limited to the preparation of symmetrical porphyrins.
EtO,
...
w
89
H* A much more versatile method introduced by MacDonald and his colleague~ ~ in~Canada is the mild acid-catalyzed condensation of 5,Y-diformylpyrromethancs (91) with 5,5'-unsubstituted pyrromethanes (92). The best
92
94
6. Porphyrins from Mono- and Dipyrrolic Precursors
169
conditions for the reaction involve the use of a very dilute solution of hydriodic acid in acetic acid and the intermediate porphodimethene 93 formed is oxidized to porphyrin 94 by aeration in a buffered solution. In the prcparation of uroporphyrins-11, 111, and IV, yields of up to 65% of porphyrin were obtained,2ebut other group^^^,^^ have since applied the method in syntheses of a rhodoporphyrin,lBcoproporphyrin-11, and etioporphyrin-11, among others. Preliminary investigations” of a route to porphyrin-a, which were aimed at applying the MacDonald method, have also been reported. However, the MacDonald method still suffers from the inherent limitation that one of the two pyrromethanes must be symmetrical, or otherwise two porphyrins could be formed. This limitation has been overcome in one notable example, the Woodward synthesis of ch10rin-e~’~ (constituting a formal total synthesis of chlorophyll-a); in the crucial stage two unsymmetrical pyrromethanes are coupled together in a unique manner, by means of an intermediate Schiff’s base. MacDonald has also condensed a 5,5’-di-unsubstituted pyrromethane (95) with a 5,5’-di(bromomethyl)pyrromethene (96) to give a 33 % yield of porphyrin, isolated as the hexamethyl ester 97.28No other porphyrins were
A
97 95
Anrc = CH,CO,Me; PNO= CH,CH,CO,Me
formed in this reaction, thus further demonstrating the utility of pyrromethanes in porphyrin synthesis. C. From Pyrromethenes
The earliest examples of the classical Fischer s y n t h e s i ~of~ porphyrins ~~~~ involved the self-condensation of 5-bromo-5’-methylpyrromethenes(98a) in a melt of succinic or tartaric acids at temperatures in the range 160-200”. In
170
The Total Synthesis of Pyrrole Pigments
this way Fischer obtained good yields of symmetrical porphyrins such as etioporphyrin-I (99a) and coproporphyrin-I (99b), even better yields could be obtained if the corresponding 5’-bromomethylpyrromethenes (98b) were
R
Me
Me Br
Me
CHzX
hie 98a X = H 98b X = Br
R
99a R = Et 99b R = P
heated at 100” in formic acid.* This type of synthesis is largely limited to centrosymmetrical porphyrins because condensation of two different pyrromethenes can lead to three different porphyrins, two by self-condensation and a third by cross-condensation. However, i n many cases, the porphyrin arising from cross-condensation has been successfully separated from the mixture.76 A more general approachasbsubsequently developed by the Munich School involved the condensation of 5,S-dibromopyrromethenes (100) with 5 3 ’ dimethyl- or 5,5’-di(bromomethyl)pyrromethenes (101) in an organic acid melt. Fischer’s synthesis of deuteroporphyrin-IX (102), a crucial intermediate in his synthesis of protoporphyrin-IX (103) and hemin, is a classic example
-+-+Br
NH H V 100
XH,C
lOla X = H lOlb X = Br
Br
v Fe(OAc),
___f
Me
Hemin
Me
P
P 103
Scheme 3
171
E iM *e
Br NH H N
Mc$YYJE,
CH,Br
>--.
+
Me
P
MC
P
Br
R 104; 1%
+ Etioporphyrin R = H, Br
Lit
CH,Br
V C
Br
Et d
Me P
Mc
Br BrCH,
c-0, Et
Nli
N
Mc
106b
M
HN
K R
=
N
R’ = Et
= V ; R’ =
N:
:
HN
P
Et
Meso-V (main product)
El
106c R = R’ = V. (106a yield << 1 %)
172
c NH a
Me M e CO, H
I’ 106a
N
+ trace of another porphyrin
6. Porphyrins from Mono- and Dipyrrolic Precursors
173
(Scheme 3).2*76 Acetylation of the deuteroporphyrin-IX was carried out on the iron complex, because Friedel-Crafts type reactions are unsuccessful with free porphyrins, presumably owing to formation of dication species. Apart from symmetry considerations, other limitations to the usefulness of the pyrromethene fusion method for porphyrin synthesis are that labile substituents may not survive the drastic experimental conditions, and, moreover, yields may be extremely low. However, in spite of all these difficulties, the Fischer method has been of immense utility in the synthesis of a wide variety of different porphyrins; for example, in the work leading up to the determination of the structure and synthesis of hemin,'O Fischer synthesized the four etioporphyrins, and 12 of the 15 isomeric mesoporphyrins.a The total synthesis of many of the degradation products of chlorophylls a and 6 , such as pyrroporphyrin-XV" (104), phylloporphyrin-XV7* (105), rhodoporphyrin-XV8 (106) (Scheme 4) enabled the Munich School to deduce their structures almost completely.
Me$Me
+
hNH$ : 'H N
Me
Br
P
CH,
+:M Q :e
Me
N
Me
Br
P
HN R
Me 105; 3.5 %
R = H. Br
+
M
Et
Me
Me Me
Me
e N Ha H N N;
Et
P
H
Meso-V
Scheme 4
;
174
The Total Synthesis of Pyrrole Pigments
In recent years perhaps the most notable applications of the Fischer method have been the achievements of MacDonald’s group in Ottawa. Their syntheses of all four u r o p ~ r p h y r i n s ~and ~ ~ *of~ *the ~ ~corresponding coproporphyrinsao showed definitively that the naturally occurring isomers were exclusively of types I and 111. The identity of the compounds produced by the Fischer method, with those they also obtained from pyrromethanes, and with the natural products was rigorously confirmed by m.p., X-ray, and chromotographic comparisons. (As MacDonald has pointed out, m.p. or X-ray comparisons are not always sufficiently reliable by themselves for confirmation of identity owing to the polymorphic behavior of many porphyrins;80*81however, it must also be said that the massive edifice of porphyrin structural chemistry erected by Fischer on m.p. comparisons has well withstood the test of time and the impact of the more sophisticated spectroscopic and chromatographic techniques now available.) Other important applications of the Fischer method by the Canadian group include the synthesis of a number of degradation products82 of heme-a, and of the Chlorobium chlorophylls (650) and (660) which have considerably helped in structural studies.81 For example, resorcinol fusion of heme-a affords a “des-methyl deuteroporphyrin” and MacDonald and his colleagues therefore synthesized all four possible des-methyl analogs of deuteroporphyrin-IX. The 8-des-methyl compound 107, synthesized as shown, was identified with
-
H
MC
methyl succinic
acid melt ZOOQ
-MC
MC
P
P
P
107; 2 %
P
+ other porphyrins
the resorcinol fusion product, and this provided good evidence for the location of the three labile groups i n porphyrin-a, whose currently disputed structures (108) are shown here. The structures of the Chlorobiuni chlorophylls (650) (109) have been established beyond reasonabledoubt by degradative methodss3and by MacDonald’s synthesis of the derived pyrroporphyrins (11I ) from pyrromethanes.81
6. Porphyrins from Mono- and Dipyrrolic Precursors
0M
R=
!
Mc
Mc
175
Mc
Me
{ ox
Me
Mc
Me
Me Me
Me
Me
IC
Me
or{
108
However, there is much less certainty about the C/ilorobiurnchlorophylls(660). The currently accepted structurese4(110) are based on degradation to maleimides and to the corresponding phylloporphyrins (112) as well as mass Mc M~@$ M e N Ha N l :
I
Me
H CHz
I
PMg,
HN
Me
R2
P
H 0
CH,
I
60, fnI.ncsyI 109
But PP
Et Et Me
PP
Me Me
1 2 3 4
Buf Et
6
Et
5
Et
H 111
176
The Total Synthesis of Pyrrole Pigments
Me
I
HCOH
Me
P
112
H
CO, rarnesyl I10
Fraction
R1
1 2
Bu' Bu'
3 4 5 6
Pr"
Pr" Et El
K2
Et Et Et Et Et Me
K Et Me El
Me Me Me
spectral determination^.^^ MacDonald's group have since synthesizeds1 the phylloporphyrins from fractions 5 and 6 by the pyrromethene method; unfortunately, a similar synthesis of the phylloporphyrin from fraction 3 gave very low yields and comparisons with naturally derived material were inconclusive. There is still some doubt about the location* and nature of the nieso-substituents, for Mathewson, Richards, and Rapoportss have reported N M R evidence favoring the a-position (rather than the 8-); the Liverpool group's results have also cast some doubt on the existence of any mesoethylated Chlorobiirni chlorophylls.
D. From Pyrtoketones Until recently pyrroketones were little more than chemical curiosities, but ~ ~ found since 1966 their tetrapyrrolic analogs, the a- and b - o ~ o b i l a n e s ,have
* Reductive C-methylation (cf. 25 and 26) has recently confirmed the siting of the mesoalkyl groups at the y-positions (S. F. MacDonald, private communication).
6.
Porphyrins from Mono- and Dipyrrollc Precursors
*
177
extensive use in the synthesis of porphyrins and oxophlorins. However, the direct synthesis of oxophlorins from pyrroketones has also attracted attention recently and Clezy's group in A ~ s t r a l i a ~ *prepared - ~ ~ a number of oxophlorins (115) and acetoxyporphyrins (116) from 5,5'-diformylpyrroketones (113) and 5,5'-unsubstituted pyrromethanes (114a) or their dicarboxylic acid analogs (114b). 0
CHO
OHC
113
€1__3
air
a 115
.1
11411 R = I1 114b R = COZH
meso-Unsubstit uted porphyrin
-4 I )2)H,/Pd-C air. I*. or DDQ
116
The diformyl pyrroketones are usually prepared by direct oxidation of the corresponding 5,5'-dimethylpyrroketones with lead t e t r a ~ e t a t e . * *The ~~~~~~ cyclization is of course analogous to the MacDonald method with pyrromethanes, but the formyl groups which are to form the bridging carbon atoms must be situated in the pyrroketone moiety, because the om-function in a pyrroketone deactivates the 5- and 5'-positions toward electrophilic attack.sQ The primary products of the reaction are oxophlorins (115) (oxyporphyrins), but these are usually converted directly to the corresponding mesoacetoxyporphyrins (116); it has been shown by other workers that the acetoxy substituent may be removed by hydrogenolysis and the porphyrinogen formed can be reoxidized to p~rphyrin.~" Like the analogous MacDonald synthesis from pyrromethanes, the Clezy procedure is, however, somewhat limited in
The Total Synthesis of Pyrrole Pigments
178
scope for preparing porphyrins because one of the two components 113 and 114 must be symmetrical, or two isomeric oxophlorins will be formed. On the other hand, i t represents a relatively rapid approach to macrocycle synthesis, and has considerable potential, for a wide variety of oxophlorins have now been synthesized including a number with electronegative substituents (e.g., bromine,92 a ~ e t y l , and ~ ~ , ethoxycarbonyls3). ~~ However, perhaps Clezy’s most noteworthy achievement has been his recent synthesiss4 of or-benzyloxyand a-acetoxy-protoporphyin-1X diniethyl esters (119a and 119b, respectively); 4-acetyl-a-hydroxydeuteroporphyrin-IX* (117) was first prepared from a diformyl monobromomonoacetyl pyrroketone, and then acetylation of the acetoxy derivative with acetic anhydride (or ether/acetyl chloride) in presence of stannic bromide gave the 2,4-diacetyl-a-acetoxy deuteroporphyrin-lX 118. Hydrolysis to the corresponding oxophlorin followed by borohydride reduction afforded the bis(hydroxyethy1) oxophlorin, which was dehydrated and benzoylated by treatment with benzoylchloride to give the desired a-benzoyloxyporphyrin 119a.The corresponding a-acetoxy porphyrin 119b was obtained in a similar manner; however, attempts to hydrolyze either compound were fraught with problems, but passage through
+
f
e
,
NH HN
h
HN Me
Mc I’
$
-
c
I17
M
€IN
Me
P
P
P
118
OK Me
HN Me
Me
P 119a
M MC f
Z
I1 N 3 ,
P
R = COPh
120
119b R = Ac
* For systematic nomenclature purposes as their hydroxyporphyrin tautomers.
i t is convenient
to name and number oxophlorins
7. Porphyrin Synthesis from Open-Chain Tetrapyrrolic Intermediates
179
a basic alumina column gave a solution of the unstable blue-green “aoxyprotoporphyrin-1X” 120. Owing to shortage of material, 120 was not fully characterized, although Clezy and LiepaD4were able to demonstrate the conversion of its ferric complex by aerial oxidation and subsequent hydrolysis to a blue product which was identical with biliverdin dimethylester in its visible spectroscopic and TLC behavior. This result is of great interest in relation to the probable metabolic route from heme to bile pigments. 7. PORPHYRIN SYNTHESIS FROM OPEN-CHAIN TETRAPYRROLIC INTERMEDIATES
A.
From a,c-Biladienes
In one variant of the pyrromethane fusion method, Fischer and S~hormiiller’~ condensed the 5-bromomet hyl pyrromet hene 121 with the 5-u nsu bsti tu ted pyrromethene 122 to give a mixture of etioporphyrin-I 124 and pyrroporphyrin-XVI11 125. These were easily separated because 125 has a side-chain
Me
Et
CH,Br
Ur 121
............. f
H
Me
Mc 122
123
Me$ HN f YJ M kt e f F f & J E t Et
Me
Et
Me 124
H --
P Me
Me 125
The Total Synthesis of Pyrrole Pigments
180
carboxylic acid group, and moreover a third porphyrin is not formed in this case because the pyrromethene 122 does not carry a 5-bromosubstituent. 'The intermediate tetrapyrrole formed en route to pyrroporphyrin-XVIII must presumably be the a,c-biladiene 123, and a logical extension of the Fischer synthesis was clearly to prepare apbiladienes of this type under mild conditions and then cyclize to porphyrin in a second step. This has now been achieved by members of the Nottingharn school,14who showed that 5-bromo5'-bromomethylpyrromethenes (126) could be coupled with a 5-unsubstituted-5'-methylpyrrornethene (127) at room temperature in presence of stannic chloride. Treatment of the reaction product with methanolic hydrogen bromide gave the relatively insoluble a,c-biladiene hydrobromides (128), which were then usually cyclized directly to porphyrins (129) without further purification by heating in boiling o-dichlorobenzene for a few minutes. Some
Br 126
-
Rcw 1) SnCI,
2) HBr
kH
HN
127
/
128
J
hot o-C,H,CI,
P 129
R = H or Alkyl
130
7. Porphyrin Synthesis from Open-Chain Tetrapyrrolic Intermediates
181
twenty or more porphyrins have now been prepared by this method, yields of up to 80% being obtained in the first stage and up to 90% in the second ~ t a g e . ~A~useful * ~ ~modification - ~ ~ ~ of the reaction conditions for the cyclization, which gives improved yields in some cases, is to dissolve the a,cbiladiene in dimethyl sulfoxide and pyridine at room temperature with exclusion of light.g8After about two days porphyrin formation is complete, and mesoporphyrin-IX (130), for example, has been prepared in 78% yield in this way. Lower yields of the intermediate a,c-biladienes and of the porphyrins are often obtained in the synthesis of meso-substituted porphyrins or those containing electron-withdrawing groupsQQ(e.g., C0,R). However, the yields of rhodoporphyrin-XV (106) (31 %) and of y-phylloporphyrin-XV (105) (29 %), two key degradation products from chlorophyll, still represent a very substantial improvement on the earlier Fischer syntheses. A number of
H
P
Me
P
131a R = H 131b R = CH,CH,NH,
Pemptoporphyrin
1
I ) FWII complex 2) ClrCHOMe/SnC14 3) -Fc
CHO I
Hofmann (R = CH,CH,NH,)
Me I
Me
J.
Protoporphyrin-IX 103
P
P 132
182
The Total Synthesis of Pyrrole Pigments
porphyrin esters, including rhodoporphyrin esters, were in fact synthesized in order to evaluate the effect of the proximity of the ester group to the position of cyclization .99 More recent examples of the biladiene method include the synthesis of pemptoporphyrin'o', chlorocruoroporphyrin*O' (132) and protoporphyrin1X102(103). Pemptoporphyrin has recently been isolated from human feces, while the heme from chlorocruoro (Spirograpliis) porphyrin is the oxygen-carrying pigment of the polychete worm Spirograpliis Spaltanz m i i . In these cases the vinyl groups were introduced by Hofmann degradation of aminoethyl side chains carried through from the pyrrolic precursors, and the formyl group of chlorocruoroporphyrin was introduced by direct formylation of a metal complex of pemptoporphyrin. An interesting feature of the behavior of a,c-biladienes is that they are readily converted into the corresponding a,b,c-bilatriene mono-salts (133), for example, on heating in dimethyl sulfoxide; the loss of proton from the b-meso-position is facilitated by an electron-withdrawing substituent in the neighboring rings. Addition of base to the dimethyl sulfoxide solution causes
H H
133
Porphyrin
Scheme 5
7. Porphyrin Synthesis from Open-Chain Tetrapyrrolic Intermediates
183
rapid conversion to porphyrin through a transient green color (the a,b,cbilatriene free base?).BBThis observation has formed the basis for a further modification of the biladiene synthesis in which solutions of the salts in dimethyl sulfoxide were treated with various one-electron oxidizing agentsand yields of up to 46% of porphyrin were obtained within 1 or 2 hours.lo3 The overall mechanism suggested for the oxidative cyclization is shown in Scheme 5 , and it may well be that the cyclizations in hot a-dichlorobenzene or
cu++ boiling
DMF
Porphyrin copper complex
134
dimethylsulfoxide and pyridine also take place by a similar mechanism, the oxidizing agent being aerial oxygen. This mild and rapid method is of potential use in the preparation of naturally occurring porphyrins bearing labile side-chains, and another synthesis103which proceeds under mild experimental conditions is the oxidative cyclization of 1’,8‘-dimethyl-a,c-biladienes(134) with an excess of a cupric salt in boiling N,N-dimethylformamide. High yields of porphyrin were obtained after heating for only 2 minutes and the effects of the nature and concentration of the cupric salt were also studied. This method has considerable potential, particularly in the synthesis of more symmetrical porphyrins, because the 1‘,8’-a,c-biladienes (134) can be readily prepared by condensation of a-free pyrroles with 5,5’-diformylpyrromethanes.Attempts to adapt this method to the synthesis of porphyrins via a,c-biladienes bearing
184
The Total Synthesis of Pyrrole Pigments
acetyl substituents in the terminal rings were less satisfactory, very low yields being obtained in the cyclization step.lo4 In general it can be said that the Johnson two-stage version of the pyrromethene fusion synthesis is such a very great improvement on the classical Fischer method that it can be regarded as a new method in its own right. Its versatility is shown by the ability to synthesize not only unsymmetrical alkylated porphyrins but also those bearing labile substituents such as alkoxycarbonyl or meso-alkyl substituents. As with the other currently available porphyrin syntheses, however, the introduction of vinyl, formyl, and acetyl groups must usually be effected at the porphyrin stage. Apart from these minor limitations the only other problems encountered lie in the synthesis of the intermediate pyrromethenes and their coupling to biladienes. In some instances it has proved impossible105to brominate the 5-bromo-5'methylpyrromethene 135a ro the 5'-bromomethyl derivative 135b. In two
135a X = H 135b X = Br
other cases the stannic chloride complex of the brominated apbiladienes could not be decomposed in acid without destroying the biladienes.lol Direct cyclization gave low yields of the stannic porphyrins, from which it was difficult to remove the metal, and thus protection of vacant p-positions in the intermediate pyrromethenes and their precursors seems to present difficulties. It should also be mentioned (although it does not directly fall within the scope of this review) that a,c-biladienes have proved to be useful intermediates in the synthesis of azaporphyrins, and especially of corroles and tetradehydrocorrins, whose carbon skeleton is present in vitamin
B. From a-Oxobilanes In contrast to the Fischer porphyrin synthesis, and the later modifications involving isolation of intermediate a,c-biladienes, the biosynthesis of porphyrins presumably involves stepwise coupling of four porphobilinogen uni 1s followed by cyclization to uroporphyrinogen-I I1 ; the bridges bet ween the rings in the latter and also in its open-chain tetrapyrrolic precursor are all methylene groups. Unfortunately, however, such tetrapyrroles (bilanes)
7. Porphyrin Synthesis from Open-Chain Tetrapyrrolic Intermediates
185
cannot be prepared in the laboratory in a completely stepwise fashion; moreover, when prepared in other ways they are not only very susceptible to oxidation but readily undergo acid-catalyzed rearrangements, so that attempts at cyclization lead to mixtures of porphyrinogens. This was in accord with Mauzerall’s finding that the porphyrinogens themselves (prepared by reduction of porphyrins) readily isomerize in acidic media. In consequence of these findings the Liverpool group turned their attention to the synthesis of open-chain tetrapyrroles partially stabilized by carbonyl linkages between the rings.lb The initial stages of this work relied upon the development of a new pyrroketone ~ y n t h e s i s(described ~~ earlier) and the required a-oxobilanes (138a) bearing protective benzyl ester groups in both terminal rings (at the 1’- and 8’-positions) were prepared by coupling 5’pyridiniummethyl pyrroketones (136c) with the sodium salts of pyrromethane-5-carboxylic acids (137).Ioe The pyridiniummethyl pyrroketones 136c were readily available from the corresponding 5-methylpyrroketones 136a by chlorination (with sulfuryl chloride or r-butyl hypochlorite) followed by treatment with pyridine. The a-oxobilane-l’,8’-dicarboxylicacids 138b prepared by hydrogenolysis of the benzyl ester groups could not, however, be cyclized to porphyrins by condensation with a variety of one-carbon units owing to the relative inertness of pyrroketones toward electrophilic attack (discussed earlier). The highly crystalline a-oxobilanes were therefore reduced with diborane to the corresponding bilanes 139a and hydrogenolyzed to the di-acids 139b. (The carbony1 function in pyrroketones is fairly resistant to hydrogenation or to borohydride reduction owing to its highly polar character, but this in turn makes it readily reducible by diborane, a highly electrophilic reagent.) The bilane dicarboxylic acids 139b, give mixtures of porphyrins on cyclization and aeration. However, oxidation (with one molar equivalent of r-butyl hypochlorite), followed by cyclization with trimethyl orthoformate in dichloromethane in presence of trichloroacetic acid, and aeration, gave isomerically pure porphyrin; in the first example studied, mesoporphyrin-IX dimethylester was isolated by chromatography and shown to be completely homogenous by thin-layer chromatography, NMR, and mixed m.p. comparisons with naturally derived material.loe It was concluded that the major oxidation product from the bilane di-acid 139b was the corresponding b-bilene 140 and that this had then cyclized to porphyrin, without rearrangement owing to the protective effects of the unsaturated linkage and of the terminal carboxyl in preventing acid-catalyzed cleavages of the central and the end rings respectively. Moreover, the terminal carboxyl groups could be expected to protect (at least partially) the a- and c-methylene groups of the bilane from oxidation, and even if formed the a- and c-bilenes would not cyclize to porphyrin under the mild conditions of the reaction.Io6
138a R = CH,Ph 138b R = H
137
139a R = CH,Ph 139b R = H
140
I
V
Et
Me
Me
P 141
186
P
(P = CH2CH2C02Me)
7. Porphyrin Synthesis from Open-Chain Tetrapyrrolic Intermediates
187
Although the overall conversion of a-oxobilane 138a to porphyrin 141 involves five stages, yields of porphyrin were in the 25-30% range, and there was no evidence of formation of mixtures. It was not necessary to isolate the various intermediates, but each of the individual stages could be followed and checked spectroscopically. The method has been applied to the synthesis of a number of other porphyrins, including coproporphyrins-I11 and IV,lo7 (isopemptoprotoporphyrin-IX,loB and 2-vinyIde~teroporphyrin-IX~~~*~~~ porphyrin) methyl esters. The free 4-position in the latter was protected by a bromine substituent until the porphyrin stage, while the vinyl group was introduced by transformation of an acetoxyethyl substituent (which had been carried through from the pyrrole stage). Earlier examples of the synthesis suffered from the limitation that the pyrromethane carboxylate 137 was derived by partial hydrogenolysis of a symmetrical dibenzyl ester. This disadvantage has now been overcornello by use of a pentachlorophenyl ester (as an alternative protecting group) which can be removed by mild alkaline hydrolysis while retaining a nuclear benzyl ester group. The completely unsymmetrical mesoporphyrin-XI dimethyl ester (141) has now been synthesized in this way,110as has protoporphyrin-IX dimethyl ester.lO* C. From 6-Oxobilanes and Oxophlorins A logical corollaryse to the a-oxobilane method was to synthesize the analogous b-oxobilanes and study their cyclization. Coupling of the phosphoryl chloride complexes of the pyrromethane amides (142) with the 5unsubstituted pyrromethane 143a followed by hydrolysis of the intermediate imine salts (144a) afforded the 6-oxobilanes (144b) in good yield. Hydrogenolysis of 144b followed by cyclization with trimethyl orthoformate in dichloromethane in presence of trichloroacetic acid affords “oxomethenes” (145) which, upon aerial oxidation, give the oxophlorinsaa (“oxyporphyrins”) (146) in up to 70% yield from the b-oxobilanes (144b). The oxophlorins can be isolated as dcep blue crystalline solids, but in solution they are unstable and undergo photooxidation to form ill-defined red pigments.”‘ For conversion to porphyrins it is usually more convenient to treat the crude reaction product with acetic anhydride and pyridine. The resulting meso-acetoxyporphyrin (147) can then be catalytically reduced to the meso-unsubstituted porphyrinogen and reoxidized to form the corresponding meso-unsubstituted porphyrin 148, preferably by aerial oxidation or with DDQ.98(Dilutesolutions of iodine have also been used, but if there are any free 8-positions in the porphyrinogen, they may be iodinated and iodoporphyrins will be formed.) The earliest examples of the use of this method were the syntheses of mesoporphyrin-IX,Oacoproporphyrin-111 and protoporphyrin-1XloBmethyl
w \
PIiCH,O,C'
CONMe,
PliCH ,0,C
143a R = H
143b R = COIBu'
x
+ 144a X = N M e , 144b X = O
t -
146
188
147
-f+& AH HN
145
148
0
7. Porphyrin Synthesis from Open-Chain Tetrapyrrolic Intermediates
189
esters; as with the a-oxobilane route the 5-unsubstituted pyrromethane 143a was derived from a symmetrical dibenzyl ester by partial hydrogenolysis, followed by decarboxylation. More recently this limitation has been overcome by use of pyrromethane 5-benzy1, 5’-t-butyl esters (143b).l1°In this way completely unsymmetrical pyrromethanes were prepared and the t-butyl esters removed selectively by hydrolysis and decarboxylation with neat trifluoroacetic acid. The resulting 5-unsubstituted pyrromethane benzyl esters (143a) were then used directly without purification for coupling with pyrromethane amides (142). The vinyl side-chains of protoporphyrin-IX were introduced via acetoxyethyl side chains (as in the a-oxobilane route), which were modified at the porphyrin stage.lo8 Unlike the crystalline a-oxobilanes, most of the b-oxobilanes so far prepared have only been obtained as “foams” or gums, but the homogeneity of these intermediates after chromatographic purification has been shown by TLC and N M R (and to a lesser extent by mass spectrometry, since their high molecular weights render them unstable and molecular ions are not often observed). A useful modification of the work-up procedure is to chromatograph the intermediate imine salts (144a) before hydrolysis; starting materials and neutral impurities may be readily removed by elution with benzene and ethyl acetate, leaving the polar imine salts on the column, from which they can then be stripped with methanol and ethyl acetate. The method has considerable potential for synthesis of a wide variety of porphyrins, and in addition to the examples already mentioned it has been applied to the synthesis of the methyl esters of 4-vinyl-deuteroporphyrin-IX109 (pemptoporphyrin) (150), rhodoporphyrin-XV (106a) and its 2-vinyl (106b) and 2,4-divinyl (106c) analog^,^^^"^^ as well as intermediates required in approaches to the synthesis of porphyrin-a.Il4 The 4-vinyl deuteroporphyrin dimethyl ester was shown to be identical with the dimethyl ester of pemptoporphyrin, a recently isolated fecal metabolite, by mixed m.p., mass, and NMR spectral comparisions. The synthetic 2-vinyl isomer prepared by the a-oxobilane route was shown to be different from the natural material by mixed m.p. comparisons as well as by differences in NMR s p e ~ t r a . ~ ~ ~ , ~ ~ ~ The intermediate acetoxyethylporphyrin 149a (en route to pemptoporphyrin 150) was also formylated via its iron complex, and the product 149b was converted by the sequence shown in Scheme 6 to 2-formyl-4-vinyl deuteroporphyrin-IX dimethylester 132. The latter was identical in all respects with the porphyrin from chlorocruoroheme (spirographis heme), the oxygencarrying pigment of the polychete worm Spirographis spallanzanii and with one of the two formyl porphyrins obtained indirectly by photooxidation of protop~rphyrin.~~~ The synthesis of rhodoporphyrin-XV dimethyl ester (106a) is outlined in
c
M
'
c
W C l i ? CH ,OAc
N H IiN 1'llc'I
J
!
T
1 @.?C;
f
'CON Me,
$
-
e
M
CH,OAc
Me
Me p \I,.
plh
J"._iM;.
v
149a R = H 149b R = CHO
1
..* 0-CH
e
M Me
Me pile
p>lo
-
Me
CHO I
-
\ J
1
I ) hydrolysis 2) mcrvl chloride 3) I-butoxide
MC
p,lc
132 Scheme 6
190
CH,/ C H . , O h
- CH,CH,CO,Me
7. Porphyrin Synthesis from Open-Chain Tetrapyrrolic Intermediates
191
Scheme 7, and the corresponding mono- and divinyl analogs were prepared
NH HN e ; CONMe, J ? - Y+ jeM
PhCH,O,C PhCH,O,C R 2M *e
Mc
Me
p Jlc
CO, R
PLIC
Mejs:" N
NH
Mc
N
t+-
N
co,R
HN
HN
p ,410
p .\I<.
C=O
/
CH,
C.O,Me 152
CO, R
a R1= Rz = Et b R1=V;R2=Et c R1=R2=v 106
Scheme 7
in the same way utilizing acetoxyethyl side chains as precursors of the vinyl g r o u p ~ . ~ ~Modification ~ ~ 1 ~ 3 of the nuclear ester function at position-6 as shown below then affords the corresponding 6-/?-ketoesters113(152), whose magnesium complexes are .of great interest as possible intermediates in the biosynthesis of chlorophylls. Indeed, the capacity of oxophlorins (such as Por-6-COZR --f Por-CO,H -+ Por-COC1 ---t /Co2Me
Por-COCH
\
-+ Pot-COCH,CO,Me
C02Bu'
192
The Total Synthesis of Pyrrole Pigments
151) to undergo hydrogen exchange at the meso-position opposite the 0x0-function"' (see below) has made it possible to synthesize tritium-labeled rhodoporphyrins (106) and their /3-ketoester analogs (152). Furthermore, 14C labels can be introduced readily into the ketoester side chain by use of appropriate malonate derivatives, and consequently single- and doublelabeled biosynthetic experiments are now feasible. Preliminary resultslle have shown that an isolated chloroplast system obtained from bean leaves can convert meso-tritiated magnesium protoporphyrin-IX (prepared by the 6-oxobilane route) into chlorophyll-a, and the stage is thus set for definitive experiments on the role of porphyrin /3-ketoesters in chlorophyll formation. I n vitro experiments113have already shown that oxidation of the magnesium
Et
-
MC
C0,Me 153
complex of ketoester 152a with iodine in methanol in presence of carbonate effects cyclization to the pheoporphyrin derivative 153. A further recent example117of the 6-oxobilane method was the synthesis and proof of structure of Hardcroporphyrin, a monovinyl tricarboxylic acid porphyrin isolated from the Harderian glands of the rat. Both the dimethyl ester 154c and its isomer 154d were synthesized, and 154c was shown to be identical with the natural product by m.p., N M R spectral, and countercurrent comparisons. In addition to its usefulness in the synthesis of porphyrins, the 6-oxobilane route also brings with it the added bonus of making rationally synthesized unsymmetrical oxophlorins readily available. These compounds were originally prepared by Lemberg, Fischer, and their co-workers by oxidation of the pyridine complexes of Although this method leads to mixtures, there is no doubt about the spectroscopic identity of the chromophores obtained with those prepared more recently by ring s y n t h e s i ~ ~and ~*~~' with other more symmetrical oxophlorins recently obtained directly from diformyl pyrroketones.88-95
7. Porphyrin Synthesis from Open-Chain Tetrapyrrolic Intermediates
R'
193
Me
/
154c or 154d
154a or 154b 154a R' = CH,CH,OAc; Ra = PMe 154b R1 = pare; R2 = CH,CH,OAc 154c R' = V; Ra = Pnre 1546 R ' = Phic; R2 = V
Interest in oxophlorin synthesis has arisen not only because of their use as intermediates in porphyrin synthesis, but more importantly because it has long been thoughtlZ0that the a-hydroxy derivative of hemin is an intermediate in the breakdown of hemin to bile pigments. The red metal complexes of oxophlorins (or "oxyporphyrins" as they used to be known) must exist in the 401,588, and tautomeric meso-hydroxy form unlike the blue free bases (A,,, 635 nm) in which the oxygen function is clearly in the om-form"' (as shown by their visible spectra and the carbonyl absorption at 1560 cm-l, which is closely similar to that observed for simple pyrroketones). The N M R spectra of the free bases are somewhat ill-defined owing to partial free radical charThe free acter, but the spectra of the salts are quite well radical character has been attributed to a low energy triplet state, and Bonnett has recently shown that the free electrons are largely associated with the methine bridges rather than the peripheral positions.lal The oxophlorins are very strong bases, being readily converted into their monocations (A,, 407,500, 537, 584) even by acetic acid; stronger acid is, however, required to 416, 560, 615 nm).ll1 convert them into their violet red dications (A,,, The monocations readily undergo exchange of the meso-proton opposite to the oxo-group in mildly acidic media and this provides a useful method of tritium 1abeling.l'' The ferric complex of a-oxymesoporphyrin-IX. (155) labeled in this manner (asterisk in formula) has recently been shown in the rat to be degraded to mesobilirubin and this provides good circumstantial evidence for the involvement of the a-oxy derivative of hemin in the normal
194
The Total Synthesis of Pyrrole Pigments
AH IiN
0
Monocation
Free base
N,0-Dication
N,C-Dicalion
breakdown of haem to bile pigments122(see also below). The ferric complex of the p-oxy isomer is not, however, degraded to the corresponding mesobilirubin in uiuo. Et
OH
Me
In uluo ___+
Me
Mc
P 155
N11
Me P
P
Mesobilitubin-IXor
D. From 6-Bilenes
In the synthesis of porphyrins from a-oxobilanes the penultimate stage is the cyclization of a b-bilene dicarboxylic acid. A more direct r ~ u t e * ~to* such '~~ b-bilenes is the condensation of an a-formylpyrromethane (157) with an
7. Porphyrin Synthesis from Open-Chain Tetrapyrrolic Intermediates
195
a-unsubstituted pyrromethane (156; R = H) or the corresponding acarboxylic acid (156; R' = C02H). Initially attempts were made to utilize b-bilene dibenzyl esters (158a), but removal of the benzyl groups by hydrogenolysis could not be accomplished without concomitant hydrogenation of the methene linkage. However, the di-t-butyl esters (158b) were readily cleaved by cold trifluoroacetic acid and the resulting 1',8'-di-unsubstituted
*3 R'
RO&
156
____+
NH H N
157a R = CH,Ph 157b R = But R' = H, CO,H, or CO,Bu'
158a R = CO,CH,Ph 158b R = CO,But 1 5 8 ~R = H
1
159
b-bilenes (158c) could by cyclized to porphyrin 159 by treatment with trimethyl orthoformate in dichloromethane in presence of trichloroacetic acid. Porphyrin formation was complete in a few minutes and, for example, the dimethyl esters of mesoporphyrins-IV, X,and XI11 were prepared in 39, 47, and 57 % yields, respectively, from appropriate b-bilenes. Later attempts to synthesize rhodoporphyrins (i.e., porphyrins bearing nuclear carboxylic ester substituents) by cyclization of b-bilenes bearing a methoxycarbonyl group at position-3 (i.e., in the B-ring) gave rise to mixtures of porphyrins.*5On the other hand, a similar synthesis involving cyclization of a b-bilene bearing a 2-methoxycarbonyl group (i.e., in the A-ring) gave a moderate yield of a rhodo-type porphyrin; this was isomerically pure and
162
196
163a 163b
Scheme 8
R =Me R = Et
7. Porphyrin Synthesis from Open-Chain Tetrapyrrolic Intermediates
197
uncontaminated with other porphyrins, although cyclization was much slower, as expected, owing to deactivation by the nuclear ester group.86J14 The failure of the b-bilene method to give a single isomerically pure porphyrin when the methoxycarbonyl group was in the 8-ring may be attributed to the increased susceptibility of the methene carbon atom in this case toward nucleophilic attack, for example, by the terminal pyrrole ring of the intermediate decarboxylated b-bilenes. If this occurred, then a tripyrrylmethanetype intermediate would be formed and this could break down in three different ways (discussed earlier), and hence mixtures of porphyrins would be formed. More recent work has also shown that incertaincases havingcomplex unsymmetrically disposed side chains small amounts of porphyrin byproducts are obtained in cyclization of pure b-bilenes.I1* However, the method has been successfully applied to the synthesis of meso-substituted porphyrins related to the Chlorobium ~ h l o r o p h y l l sand ~ ~ *to~ desoxophyllo~~ erythroetiop~rphyrin.~~~ The meso-substituted pyrromethanes (160a) were first prepared as shown in Scheme 8 and the corresponding carboxylic acids (160b) were condensed with the formyl pyrromethanes (161c) to give the b-bilenes (162). Subsequent cyclization of b-bilenes followed the usual procedure, deesterification and decarboxylation with trifluoroacetic acid followed by cyclization with trimethylorthoformate-trichloroaceticacid and aerial oxidation. However, the yields of the meso-methyl and meso-ethylporphyrins (163a) and (163b) were only about 5 % compared with the 50% observed in the early work. The low yields may be attributable to the influence of the meso-substituent insofar as it may sterically hinder attainment of the conformation required for cyclization; alternatively, the presence of an unsubstituted b-position may be the cause, since similar results have been obtained when a /?-position has been free but there has been no meso-s~bstituent.~~~ The two synthetic porphyrins 163a and 163b and their copper complexes were compared with the phylloporphyrins derived from fractions 3 and 4 of Chlorobium chlorophylls(660). The synthetic meso-methyl porphyrin (163a) was sh0wn13.~~~ to be identical with the naturally derived porphyrin and with an earlier synthetic sample prepared by MacDonaldel (utilizing the Fischer method) by means of X-ray powder photographs and m.p. comparisons. However, the synthetic ðyl porphyrin 163b was not identical with any of the phylloporphyrins derived from the Chlorobium chlorophylls(660) and this result in combination with other evidence has cast considerable doubts on the occurrence of d-ethyl substituents in this series.lZ3More degradative and synthetic work is clearly needed, however, before a definitive answer to the outstanding structural problems can be obtained. Rapoport approached the synthesis of porphyrins by the b-bilene route independently of the Liverpool group, and this work culminated in an elegant
198
The Total Synthesis of Pyrrole Pigments
164
165
DCV = CH=C(CN),
J Et
166a R = CH-NH.HC1 166b R = CHO 1 6 6 ~ R = CHS
Mc
Me 167
synthesis of desoxophylloerythroetioporphyrin (167), the predominant porphyrin present in petroleum. This synthesis utilizes a variety of protecting groups; for example, the acid labile anisyl protecting group was used for the pyrromethane 164 corresponding to rings D and A of the final porphyrin, while the formyl group of the other pyrromethene (165) was protected initially as a dicyanovinyl group (during pyrromethane formation) and later as
8. Chlorins and Other Partially Reduced Porphyrins
199
an aldimine hydrochloride (during 6-bilene formation). The formimino mesosubstituted 6-bilene dihydrochloride 166a was hydrolyzed to the formyl-bbilene 166b and converted to the thioformyl analog 166c. Either of these two b-bilenes could be cyclized with 2.5% hydriodic acid in acetic acid followed by air oxidation to give desoxophylloerythroetioporphyrin (167) (conditions which are similar to those employed in the MacDonald method). The yield was only 6%, but this was attributed to the steric effect of the isocyclic ring. This view was supported by results obtained by applying the ac-biladiene route; the required biladiene was obtained in excellent yield, but its cyclization in DMSO-pyridine gave less than 3 % of desoxophylloerythr~etioporphyrin.~~ In conclusion therefore it can be said that the b-bilene route is useful for the preparation of isomerically pure but fairly simple porphyrins; severe constraints may occur if electron-withdrawing groups, vacant p-positions, or meso-substituents are present. It is essentially a two-stage version of thc MacDonald synthesis enabling it to be applied to the synthesis of less symmetrically substituted porphyrins; a b-bilene was also a fleeting intermediate in the Harvard synthesis72of chlorophyll. 8. CHLORINS AND OTHER PARTIALLY REDUCED PORPHYRINS
Although the chemistry of chlorophylls a and b (168) has been studied e x t e n s i ~ e l y , ~there * ~ ~ is ' ~as ~ yet no direct rational stepwise synthesis of the dihydroporphyrin (or chlorin) ring which they contain. Most of the synthetic chlorins to date have been prepared either by degradation of chlorophylls or by reduction of porphyrins. The earliest preparation of chlorins from porphyrins was Fischer's reduction of porphyrin iron complexes with sodium in amyl alcoh01.~More recently Inhoffen and his colleag~es12~ have shown that diborane reduction of octaethylporphyrin in tetrahydrofuran affords a mixture of cis- and transoctaethylchlorin (ratio 5 :1, respectively) whereas diimide in pyridine reduction is stereoselective affording the cis-isomer only;127in contrast, the sodium in alcohol reduction affords the trans-isomer. In more complex cases, it would be difficult to control the position of reduction and so these methods are of limited utility for chlorin synthesis; however, Fischer has claimed1z8 that reduction of y-phylloporphyrin-XV (105) affords the 7,8-dihydro derivative (169), presumably because reduction of the 7,8-double bond relieves the steric strain129between the y-methyl group and the 7-propionate side chain. Simple symmetrically substituted chlorins have been prepared by selfcondensation of 2-dimethylaminomethyl pyrroles in presence of ethyl magnesium bromide in boiling ~ y l e n e like ; ~ ~other ~ chlorins they can be
200
The Total Synthesis of Pyrrole Pigments
v
K
H
CO, Phytyl 168a R = Me 168b R = CHO
Me
El
I
P
Me
Et
I
Mc 105
111
H P
Mc 169
readily dehydrogenated by high potential quinones to the corresponding porphyrins.131 Simple tetrahydroporphyrin~~~~ (analogs of the bacteriochlorophylls) have also been prepared. During the course of work on the synthesis72of chlorophyll (see below) Woodwardlo and his colleagues discovered the phlorins (170),a new type of dihydroporphyrin formed as intermediates in the synthesis of chlorophyll-a. The phlorins which have additional hydrogens on nitrogen and on one of the meso-carbon bridges can be prepared by reduction of the porphyrin ring system in various ways-by addition of thiol acids, photoreduction, and other one-electron p r o c e s ~ e s . ~Their ~ * ~blue ~ * color ~ ~ ~ (A,,,ax, ~ 620 mm) is very characteristic, and they also form green monocations (Amax, 725 mm) by protonation on the remaining nitrogen; in strong acid further protonation occurs on carbon to give a “bis-porphomethene” (171)Y Phlorins are very readily reoxidized to porphyrins by mild oxidizing agents such as air or iodine; further reduction affords porphomethenes and porphyrinogens.1°
8. Chlorins and Other Partially Reduced Porphyrins
201
An interesting new synthesis of simple chlorins and their reduced porphyrin derivatives has been explored recently by the Braunschweig Oxidation of octaethylporphyrin with hydrogen peroxide in sulfuric acid or with osmium tetroxide affords the diol 172, which on treatment with acid undergoes a pinacol-type rearrangement to the chlorin, or “gemini ketone” 173. Repetition of the same process leads to di- and even triketones. This
B 170
Et
Et
171
El El
Et El
El
172
a,,,,, XCH 173
K
H
85
R = H or CO,H
H
H
2NH
36
A = CHgCOgH P = CHzCH2COzH
174
approach would, however, be difficult to apply to more complex naturally occurring chlorins, although it is potentially useful in corrin and Vitamin B,, chemistry. Porphyrinogens (174), the colorless hexahydro derivatives of porphyrins, are readily prepared by reduction of porphyrins by using a variety of reducing
202
The Total Synthesis of Pyrrole Pigments
agents such as sodium amalgam, sodium borohydride, zinc dust and alkali, and hydrogenation (over platinum). They are readily reoxidized by air, iodine, quinones, or other oxidizing agents to the parent porphyrin. Porphyrinogens can also be obtained135 by polymerization of simple pyrroles of type 85 under anaerobic conditions, but unless the substituents in the bpositions are identical, mixtures of isomers are obtained. However, enzymic polymerization of porphobilinogen 36 normally leads ~pecifically~~ to uroporphyrin-I11 except in rather rare pathological conditions where uroporphyrin-I is formed. Woodward’s synthesis of chlorin-e,, accomplished in 1961, still represents the only synthesis of a completely unsymmetrical chlorin from pyrrolic intern~ediates.~~ It involves first the preparation of an unsymmetrical porphyrin, followed by a series of further reactions in which the two additional hydrogen atoms are introduced into the D-ring in a stereospecific manner. The two pyrromethanes 175 and 176 were first prepared as shown from pyrrolic intermediates and then condensed together in an ammonium acetate/ acetic acid buffer. Under these conditions the highly reactive thioformyl group of the pyrromethane 176 formed a Schiff’s base with the side chain amino group of the pyrromethane 175, the ketonic moiety in the other pyrrole ring of 176 being insufficiently reactive under these conditions to interfere. The unstable product 177, in which the Schiff’s base served to hold the two methanes together in the desired manner, was then treated immediately with hot methanolic 12M hydrogen chloride and cyclized to the phlorin 178, presumably by condensation first to form the a-bridge and then the y-bridge. The phlorin was oxidized to porphyrin 179 with iodine, and after acetylation of the side chain amino group, further oxidation by air gave the mesoacrylic ester derivative 180. The latter could then be equilibrated with the purpurin 181 because of the steric interactions between the meso-acrylic ester and the substituents at the 6- and 7-positions in the porphyrin ring. The acetaniidoethyl side chain in the purpurin was then hydrolyzed with methanolic hydrogen chloride and subjected to Hofmann degradation to generate the 2-vinyl group. Subsequently, photooxidation cleaved the double bond in the isocyclic ring and the resulting 7-oxaloyl chlorin 182 underwent fission of the oxaloyl group on treatment with dilute methanolic alkali to afford a meso-formyl chlorin 183; this immediately cyclized with the 6-methoxycarbonyl substituent to form the chlorin lactone 184. Mild alkaline hydrolysis followed by reesterification afforded the racemic chlorin-e, (185), which was resolved by means of its quinine salt. The (+)-enantiomer was identical with that obtained by degradation of chlorophyll and the y-formyl group was then transformed into methoxycarbonylmethyl via the cyano-lactone 186 and the cyanomethyl acid 187; the optically active chlorin-e, (188) was identical with a sample derived from natural chlorophyll-a. This completed the formal total
CH,NH,
NC:\l
I
Me&'
,,
CN
HN
>-
\
CHzCl
H
+ OHC
Me+
I;':'..
H
2)NaBH.
CH, N H,
I
p
2
3
NEt
SHC
Et
-MC
Me
co\ 175
/
CH,
I
CH,
I
-oHc&
+-- H!&Et /
Et A
CO,El
C0,Mc
Y
176
203
t
C'H 1! N II I
CH,
I
CH,CO,Mc
177
178
C'O,El
CH,NHCOMe
I
204
I
CH,NH, I
NHCOMe
I
V
CHi
I
Mc
1) OH21 Hofmann
3) OJhv
Me
Me
H'
182
C0,Et C0,Me 181
Me H
M
pM" H /"H3,C0
C0,Me
- 'w 1
KOH/MCOH
KOH/MeOH
Me
Me,,iii'.'
CHOC0,Et
Me0 183
184
V
Me
185
1
Zn/HOAc
205
206
l h e Total Synthesis of Pyrrole Pigments
t -
1
1
CN
CO,Mc 188
Chlorin-e,
synthesis of chlorophyll-a because chlorin-e, had been previously converted1"" back into chlorophyll, although this is far from easy and has not since been repeated. Almost simultaneously with the appearance of the Harvard synthesis, Strell and K a l ~ j a n o f freported l~~ the completion of work on the synthesis of phaeophorbide-a begun many years previously in the Munich school under the aegis of Hans Fischer. The phylloporphyrin 189 was syn~hesized'~~ in very low yield from pyrromethenes and its iron complex reduced by sodium and amyl alcohol to give a product (190a) which was thought to be reduced specifically in the D-ring. Transformation of the y-methyl group into y methoxycarbonylmethyl giving 190b was then effected,I3* and the copper derivative acetylated to give the diacetyl derivative 191. Partial deacylation then gave thc 2-acetyl chlorin 192. Inhoffen has since ~ r i t i c i z e d these ~,~~~ results, and he was able to obtain only a 2-acetyl compound by modifying the experimental conditions. Borohydride reduction of the acetyl group followed by treatment of the iron complex with dichloromethylmethyl ether then gave a chlorin (193) containing the isocyclic ring. Dehydration of the 2-hydroxyethyl group and oxidation of the secondary hydroxyl function in the isocyclic ring then afforded pheophorbide-a (194). This formally completed the total synthesis of chlorophyll, since pheophorbide-a had been converted back into chlorophyll by insertion of magnesium and phytylation of the propionic acid side chain. Further work on this synthesis is probably very desirable and because of the low yields obtained in many of the stages the racemic compounds were not resolved and, furthermore, the 2-acetylchlorin (192) obtained by degradation of natural chlorophyll was used as a relay. Since thcse syntheses were completed the absolute stereochemistry of chlorophyll140 (168) and of bacteriochl~rophyll~~~-~~~ (195) have been determined. Epimers have also been shown to exist (at C-lo), and they are
Y"
I-i
I
Et
Mc li
P
Me
Mc 1908 R = Me; 190b R = CH2C02Me
189
/;!mplex
z/;mp1ex
Ac
Me
Me
I
i'0,Me
i'O,Me
191
192 I) NaBH, 2) Fe complex
+
CHCI,OMe 3) OH-
4) CH,N,
193
J
194
Chlorophyll-a 168a
207
I
CO, H
C0,Phytyl
196
195
-
R
= Et o r V
Mc
H 11 1'
CH, CO,Mc
198
I
CO,Mr 197
Chlorophyll-6 168b
t----.
C0,Me 199
208
9. Bile Pigments
209
formed in small amounts by slow equilibration of the major e ~ i m e r . " ~ The structures of the c h l ~ r o p h y l l s - c(196) ~ ~ ~ (which, it should be noted, are porphyrins rather than chlorins) and of the Chlorobiunl chlorophylls (109 and 110) have also been i n ~ e s t i g a t e d ~recently l - ~ ~ ~ but ~ ~ they have not yet been synthesized. The partial synthesis of rhodin-g, (199), a degradation product of chlorophyll-b, has recently been reported by 1nh0ffen.l~~ Electrolysis of natural chlorin-e, (188) gives the "phlorin-chlorin" 197, which undergoes photooxidation in dioxan-water to the trans-diol 198 (which is a bacteriochlorin). The latter was then transformed into rhodin-g,-trimethylester 199, and since chlorophyll4 is accessible from 199, this constitutes a formal total synthesis of chlorophyll-6, since chlorin-e, was synthesized by Woodward. 9. BILE PIGMENTS
The bile pigments are open-chain tetrapyrrolic compounds derived in nature by oxidative metabolism and ring opening of the prosthetic groups of hemoproteins.11*120 Biliverdin (200) is the first open-chain tetrapyrrole to be formed and is then rapidly reduced to bilirubin (201). With only one exception, all known bile pigments of natural origin are derived by cleavage of the
NH H N
Me=
Me
NH HN
Me
P
P
P
20 1
200
Me
NI-I HN P
00 202
P
P
Me
210
The Total Synthesis of Pyrrole Pigments
heme at the a-bridge and thus belong to the so-called IXa- series. The exception is the blue-green tegumental pigment of the caterpillar of the cabbage butterfly, which has recently been shown"' to be biliverdin-lXy (202) and is presumably formed by ring opening of heme at the y-position. The chemical transformation of porphyrins to bile pigments was first studied in the 1930s by Fischer, Lemberg, and their co-workers.11B-120 They showed that coupled oxidation of porphyrin iron complexes with hydrogen peroxide in presence of reducing agents such as ascorbic acid and hydrazine followed by hydrolysis afforded biliverdin-type pigments. Some of the earlier workers suggested that these chemical oxidations were specific and led like the natural process to IXa- pigments. However, more recent work has clearly demonstrated that an almost random oxidation of the pyridine hernochrome results in a mixture of all four possible b i l i v e r d i n ~ . Interestingly, ~ ~ ~ . ~ ~ ~ however, O'Carra146 has now shown that h iiitro coupled oxidation of myoglobin leads specifically to biliverdin-IXa, and in oitro oxidation of hemoglobin affords mainly biliverdin-IXa but with some of the IXg- isomer; on the other hand, myoglobin denatured by 8M urea gives a random mixture of biliverdins. The X-ray evidenceI4' shows that the a-bridge is the least accessible, being at the bottom of the heme crevice in the protein, whereas the y-bridge is
ocoP I1
I)(PhCO,), 2) removal of
metal
203
Biliverdins
204
9. Bile Pigments
211
exposed, and O’Carra therefore suggests that the heme-binding site must have a positive effect on the position of the natural cleavage. Intermediates in the chemical oxidation of heme and bile pigments are the oxyporphyrins (or oxophlorins as they are now known) and these can be degraded by further oxidation to biliverdins. Simple oxophlorins (204) can now be more conveniently prepared by oxidation of porphyrin metal complexes with benzoyl peroxideD5followed by hydrolysis of the resulting bcnzoates (203), but unless the porphyrin is symmetrically substituted, the products will clearly be mixtures. Another, indirect, method148 of preparing oxophlorins is by the oxidation of porphyrins with lead dioxide in acetic acid to xanthoporphyrinogens (205) followed by reduction with hydrogen bromide
205
in acetic acid, but this method has little general synthetic utility. The best methods for synthesizing isomerically pure oxophlorins, which can then be converted to bile pigments, are the ring syntheses described earlier, involving b-oxobilane intermediates, or the condensation of diformyl pyrroketones with pyrromethanes. Oxophlorins can be ruptured to give bile pigments following Lemberg’s and Fischer’s original procedures;118e119 the oxophlorin pyridine hemochrome 206 is oxidized by atmospheric oxygen to give first an intermediate verdoheme 207, which is then hydrolyzed to give
&+$
N f t N
2 c1-
O21pyridine ____t
207 206
1
hydrolysis
Biliverdins
212
The Total Synthesis of Pyrrole Pigments
the biliverdin. In this way, for example, members of the Liverpool school have synthesized mesobiliverdin-IXa'lo and mesobiliverdin-IXgll* and CIezy has prepared biliverdin-IXa (200) itself, albeit in rather low yield.D4 There is now little doubt that the long-held view120 that an oxoporphyrin iron complex is an intermediate in the catabolism of heme is correct, for tritium-labeled a-oxymesoporphyrin ferriheme was shown to be converted into mesobilirubin in the rat;122 moreover, the enzymic degradation was shown to be specific for the a-isomer, for the corresponding p-oxymesohemin was not converted into a bilirubin.122Oxophlorins are thus clearly of con siderable significance in their own right as well as being useful intermediates in the chemical synthesis of bile pigments. Useful as the foregoing methods are for bile pigment synthesis, however, they would be difficult to apply directly to compounds other than biliverdins or bilirubins for many of the bile pigments recently characterized or discovered are much further reduced; for example, stercobilinogen (208), the final, colorless product of bacterial reduction in the gut, is a dodecahydro derivative of bilirubin. Phyc~erythrobilin'~~ (209) and p h y c o ~ y a n o b i l i n ~ ~ ~
M e- ; .
NH H N Me
P
208
P 209
Me
210
(210), the red and green algal bile pigments, are at an intermediate level of
reduction and phytochrome, which appears to control the photoregulation of growth and development in plants, is closely related to phycocyanobilin,
9. Bile Pigments
213
although its precise structure (211) is still uncertain.149For bile pigments of this kind the stepwise coupling of pyrrole or reduced-type units appears Me,
H
NH H N
Me
Me
Me
P
P 211
to be the most useful synthetic method. This is the approach that was originally developed by the Munich school and successfully continued i n recent years by Plieninger’s group in Heidelberg (see below). Perhaps the main achievements in the earlier work were the confirmation ~~”~~ of the structures of bilirubin 201 and biliverdin 200 by ~ y n t h e s i s . ~Mild hydriodic acid reduction of bilirubin affords a mixture of bilirubic acid 212a and its isomer 213a and oxidation of these by alkaline permanganate leads to the corresponding orange yellow methenes, xanthobilirubic acid (214a), and isoxanthobilirubic acid (215a).163~-166 The closely related neo- and isoneoxanthobilirubic acids 214b and 215b were obtainedlS6 by resorcinol
M
eNH H H N
P
p
H N H H NE 0
R 212
a R=Me b R=H c R=CHO d R = CH,OH
t E* p
NH H N 214
R
t
213
NH H N
0 215
214
The Total Synthesis of Pyrrole Pigments
Me
P
P
P 216
Me
P 217
Me
Me
Me
P
P 218
fusion of bilirubin, and it was shown that condensation of formyl neoxanthobilirubic acid 214c with isoneoxanthobilirubic acid 2lJb gave mesobiliverdin 216.15’ Reduction of the formyl acid to the corresponding hydroxymethyl derivative 214d followed by coupling with the isoneoxanthobilirubic acid 215b similarly atTordedl5l mesobilirubin-IXa (217), whereas direct coupling1Ss of two moles of neoxanthobilirubic acid with formaldehyde gave mesobi I i r u bi n-X I I I a (2 18). The dipyrrolic acids were also synthesized directly from monopyrrolic precursors; for example, neoxanthobilirubic acid was obtained15Bby condensation of a bromopyrrole aldehyde with a n a-free pyrrole as shown in Scheme 9 , and the ring oxygen function was introduced into the intermediate pyrromethene by alkaline hydrolysis. A synthesis160 of xanthobilirubic acid involved the oxidation of 2,4-dirnethyl-3-ethyl pyrrole with peroxide to the corresponding pyrrolinone, which was then brominated in the a-methyl group and coupled with 2,4-dimethylpyrrole-3-propionic acid. The other two acids, iso-, and isoneoxanthobilirubic acids have also been synthesized by variants of the first method.160-162The synthesis of bilirubin and biliverdin was somewhat more complex owing to the necessity to protect the sensitive
9. Bile Pigments
215
I
NaOMe/H,O nt lSO‘
214b
vinyl groups during the intermediate stages.lb2This was effected by the use of urethane derivatives, conveniently generated from propionic acid side chains by Schmidt degradation. Thus condensation of the formyl methene 220 with the a-free methene 219 afforded the bilatriene 221, which by hydrolysis and exhaustive methylation was converted into biliverdin-IXa (200). Reduction of the latter by hydrosulfite then gave bilirubin (201).162 In a similar manner formyl neobilirubic acid (212c) and isoneobilirubic acid (213b) gave16s urobilin-IXa (222), and the mesobilirubin isomers “mesobilirhodin” (223) and “mesobiliviolin” (224) were also synthesized as shown.ls4 In more recent years the peroxide oxidation method for converting a-free pyrroles to pyrrolinones has been exploited by the Nottingham school,le6 particularly in relation to the synthesis of pyrrolylmethylpyrrolinones such as 226a from pyrromethanes (225). Further oxidation with manganese
C H,C H2NIiC02Et
-----+ NH
NH
Me
P
OHC 219
P
220
?HCo2Et M
c
S
/NHCo2Et y2 Me
MC
P
1’ 221
Me,SO,/OH‘
Bi lirubin-I Xcr
~
NaaS204
Biliverdin-I Xlx
201
200
21212
EtQE,
223 214c
Et
Et Me
Me 222
216
+ 215b +“Mesobilirhodin” + 213b ---+
“Mesobiliviolin” 224
9. Bile Pigments
217
dioxide or dehydrogenation with palladium-charcoal affords the corresponding methylene pyrrolinones 227a. These pyrrolinones were then hydrolyzed
Me
Me
pvrldlne
R
H 225
226
a R = C0,Et b R=H c
R=CHO
221
and decarboxylated to 226b and 22713 and formylation by the VilsmeierHaack procedure gave the corresponding aldehydes 226c and 227c. Condensation of 226b with 226c then gave the b-bilene 228,and the corresponding a,b-biladiene 229 was prepared in a similar manner from 226c and 227b.
; ; - .J$m -
NH H N
$ y $ E
NH HN
N
El
Me
Me 228
I1N
Et
Mc
Me 229
Hydrogenation of the pyrrolylmethylpyrrolinone 226a afforded the pyrrolidone 230a (obtained as a mixture of stereoisomers), and this was converted into the aldehyde 230c as before; condensation of 230b and 230c then gave the .tetrapyrrole 231. The preparation of the tetrapyrroles 228,229, and 231 provides models for the synthesis of various naturally occurring bile pigments
218
The Total Synthesis of Pyrrole Pigments
y :?J Et
H 230
R
MC
230a R = C0,Et 230b R = H 2 3 0 ~ R = CHO
Mc 231
such as i-urobilin, mesobiliviolin (224), and stercobilin. A number of other
similar model syntheses were carried by the Nottingham workers, and the use of pyrrolylmethylpyrrolinones in the synthesis of macrocyclic tetrapyrrolic structures of the dehydrocorrin typeIB5was also investigated. A variant of the peroxide oxidation procedure-direct oxidation of pyrrole and pyrromethane a-carboxylic acids-has been investigated in Liverpool ,166 for the carboxylic acids are generally more readily accessible (e.g., by catalytic hydrogenation of benzyl esters). However, mixtures of products were obtained and yields of the desired pyrrolinones were low. OH
R'CO-CHR
I
C0,Et
I - I t
R'C-CHR
-
NC COzEt
232
H
1
TR'->Po 1 R'
R
233
I
K O B ~
ArCHO/bsrc
NslNH,
H
ArCH 234
H
9. Bile Pigments
219
Perhaps the most generally useful method of preparing pyrrolinones now available is Plieninger's new ring synthesis.lB7Catalytic hydrogenation of the cyanhydrins of ,9-ketoesters with Raney nickel in acetic acid yields a diastereoisomeric mixture of hydroxpyrrolidones (232), which can then be dehydrated to the desired pyrrolinones (233). Model experiments showed that the 5methylene groups in 233 rapidly reacted with aromatic aldehydes in presence of base to give arylidene pyrrolinones (234).lB7Later it was shown that catalytic reduction of the pyrrolinones gave mainly the cis addition products, whereas sodium in liquid ammonia afforded the trans isomers; the cis compounds could also be isomerized to the trans form by treatment with potassium f-butoxide.'"* Following these model experiments on simple pyrrolinones, neoxanthobilirubic acid (214b) and isoneoxanthobilirubic acid(215b) were synthesized by
(233a)
+ Mc
214b
P
condensation of 4-ethyl-3-methylpyrrolinoneand 3-ethyl-4-methylpyrrolinone with formylopsopyrrole carboxylic acid (i.e., 2-formyl-3-methylpyrrole-4-propionic acid). Sodium amalgam reduction of 214b gave the dihydroderivate 212b, which was esterified with diazomethane (to facilitate purification) and then reduced catalytically at 11 5-120" over Raney nickel. The product was a mixture of stereoisomeric dihydroneobilirubic esters (235), which were separated by fractional crystallization. The higher melting isomer 235a (m.p. 163-165") condensed with ethyl orthoformate in presence of borontrifluoride etherate to give a stercobilin-XIIIa hydrochloride (236), m.p. 120°.168 For the synthesislea of stercobilin-IXa, the two dihydroneobilirubic acid methyl esters (235) were separately formylated with hydrogen cyanide under Gatterman conditions, but owing to the alkaline work-up the methyl substituent on the pyrrolidone ring underwent epimerization to give the trans arrangement of alkyl substituents on this ring (237). The other half of the stercobilin was similarly prepared from isoneoxanthobilirubic acid by sodium amalgam reduction followed by hydrogenation of the sodium salt under pressure. Owing to the alkaline conditions of the hydrogenation, the ethyl group
220
The Total Synthesis of Pyrrole Pigments
neighboring the pyrrolidone ring carbonyl group underwent epimerization and the product obtained was an inseparable mixture of epimeric acids (238). This mixture was condensed with each of the formyl dihydrooeobilirubic
0 I
235a 235b
P 236
Me
9 dMe M
P
238
H
e
w
OHC
e
C1-
P
P 231 237a 237b
M
239a 239b
esters 237a and 237b in acetic acid containing hydrogen bromide, giving two .stercobilins (239)which were isolated as their hydrochlorides, m.p. 208-209" and L66-169", respectively. The IR,UV, and chromatographic properties of these two compounds were practically indistinguishable from those of natural stercobilin hydrochloride.l6*The product obtained earlier by perhydrogenation of bilirubin over palladium-charcoal to stercobilinogen followed by reoxidation with ferric chloride also had very similar properties. In more recent work, Plieninger and Ruppert have synthesi~ed~''~ the dipyrrolic dicarboxylic acids 240a and 241a by condensation of 2-ethoxycarbonyl-3-ethoxycarbonylethyl-4-methyl-5-formylpyrrolewith the two pyrrolinones ,233aand 233b in 40% sodium hydroxide. (These dicarboxylic acids on heating with methanolic sulfuric acid gave neo- and isoneoxanthobilirubic acid methyl esters, 214b and 215b. respectively.) If the condensation was carried out with sodium ethoxide in ethanol then the corresponding esters, 240b and 241b, were obtained. Hydrogenation of the oxodipyrromethenes 240b and 241b over palladium on barium sulfate at 1 atm and 25"
9. Bile Pigments
221
gave the dihydrodiesters 242 and 243,respectively, in -60 % yields. Hydrogenation of 241b over Raney nickel at 120" and 130 atm gave an oily tetrahydro derivative which was fractionally crystallized to give a 60 % yield of the all-cis diester 246.The epimer 247, m.p. 129, was isolated in 10% yield from the mother liquors, and the structures of the two products were confirmed by N M R assignments. Similarly, high-pressure hydrogenation of the isomeric
a b R =- H Et
NH HN
RO,C
*eM
NH HN
0
PR C 0 2R
241
240
Et
NH H N
0 243
M
N He H N
wP
C0,Et
0 242
rEt
H-'
C'0,Et
0 246
244
PEL 0 241
CO,E1 245
diester 240b gave a mixture of stereoisomeric tetrahydro derivatives 244 and 245, which were separated by fractional crystallization, and again assigned structures on the basis of their NMR spectra. The chemical shifts of the uproton in the pyrrolidone ring were 4 . 2 ppm to lower field i n the all-cis compounds 244 and 246 than in their isomers 245 and 247,respectively, and the m.p. of the all-cis compounds were also higher than their isomers.1eg
222
The Total Synthesis of Pyrrole Pigments
Alkaline hydrolysis of three of these diesters (244,245,and 246)gave the dicarboxylic acids 248a, 249a, and 250a with epimerization of the alkyl groups on the carbon atoms neighboring the pyrrolidone carbonyl groups. These dicarboxylic acids were then resolved as their brucine salts. Thermal decarboxylation of (+) and (-) 250a and sublimation afforded the corresponding (+) and (-) monocarboxylic acids (250b) in high yield, the progress of the decarboxylation being followed by the diminution in the absorption maximum at 282 nm due to the pyrrole a-carboxylic acid chromophore. Condensation of the (+)-dihydroisoneobilirubic acid (250b) with orthoformic ester in hydrogen bromide-acetic acid at 50" then gave (+)-stercobilinllIa (251),isolated as its hydrochloride [u]"," = 2830" in chloroform, raised
248
(+) and (-) forms 250a R = CO,H
250b R = H 2 5 0 ~ R = CHO
249
2JOb
+ HC(OEt),
R
HBrlHOAc ___t
NH
X-
lik
P
P 25 1
(+) and (-) forms
+
to 3460" after three crystallizations from chloroform-acetone. (-)Stercobilin-llla hydrochloride (251)[uJbo= -3570" was also prepared in a similar manner from the (-)-acid 25O.le9 Unfortunately only the dicarboxylic acid 250a could be resolved, and so, the other two racemic acids 248a and 249a were thermally decarboxylated
9. Bile Pigments
223
and formylated (with HCN/HCl) to give the epimeric aldehydes 248c and 249c. Condensation of (-)-250b with the racemic aldehyde (24%) in acetic
Mc
acid containing hydrogen bromide gave a diastereoisomeric mixture of the two products 252 and 253.Similar condensation of the other aldehyde (249c) with (-)-dihydroisoneobilirubic acid (250b)gave a diastereoisomeric mixture of the two further products 254 and 255.168 The crude product 253 from the first reaction had an optical activity [alto = - 1040", raised by five recrystallizations from chloroform and one crystallization from methanol-ethyl acetate to -3570". The mother liquors from the crude product, and the first recrystallization, showed no optical activity, because the other product (252)is a quasi-nteso type structure. (The
e ,Et wMe. E
M H'
H''
'H
, N H HN,
'H
NH Hfj
Me
P
P 252
H.
,Et
t
Me
MC
. N H HN *H NH H k
P 253
,Et Me,
P
Me
3 H,
H' Mc
Me
P
P 254
P
Me P 255
two end rings are not identically substituted, and so it is not strictly a mesocompound). The optically active (-)-stercobilin hydrochloride 253 preparedlB9 in this way slowly decomposed above 160" and melted at 175-177", whereas the natural product was reported to decompose at 157-162'. One of the second pair of products is also a quasi-meso form (255)and the other after isolation and repeated recrystallization gave another (-1stercobilin hydrochloride (254) [a]'," =, -4200, which decomposed slowly above 158-160", and melted at 177-183 . The two synthetic (-)-stercobilins had electronic spectra and chromatographic behavior identical with natural sterc~bilin.~~~
224
The Total Synthesis of Pyrrole Pigments
However, comparison of X-ray powder photographs showed that the (-)-stercobilin hydrochloride 253 (prepared from the aldehyde 248c) and (-)-dihydroisoneobilirubic acid 250b was identical with the natural material.169Thus the relative configurations of the two end rings are now known, but their absolute configurations remain to be determined. * Plieninger and Ruppert also synthesized (+)-stercobilin hydrochloride by condensation of (+)-dihydroisoneobilirubic acid (250b) with the racemic aldehyde 248c; the crude product had [a]: = +2800" raised to +3430" by recrystallization from chloroform-acetone. The precise nature of the intermediates between bilirubin 217 and the colorless stercobilinogen 209 and the order in which they are formed in nature is still not entirely clear. (+)-d-Urobilin, for example, is thoughtlz1 to be the hexahydrobilirubin derivative 256a or 256b containing one vinyl group, while i-urobilin (257) is the corresponding octahydro analog in which both the original vinyl groups of bilirubin are reduced.l" The latter, whether
-
e
:;u,
M
NH HN
Me
Me
Mc
P
P
P
256a R = Et; R' = V 256b R = V ; R' = Et
251
-NH H N
Me
Mc
P
P
258
P
P
259
* Added in proof: Degradations of the natural and synthetic products to rnethylethylmaleimides by Dr. Brockmann have now shown that the terminal rings have the RR and SS configurations respectively: hence the natural product has the RRS . . SRR configuration and the synthetic material thc SSS . . . SSS configuration. (Prof. H. Plieninger, private communication).
.
9. Bile Pigmenh
225
isolated from urine or feces or of synthetic origin, is optically inactive. A further reduction step affords a “half-ster~obilin,”~~~ 258 or 259. Some of these compounds (e.g., d-urobilin analogs) have now been synthesized by Plieninger and c o - ~ o r k e r s . ’The ~ ~ formyl pyrrole 260 was condensed in presence of methanolic sodium hydroxide with the protected aminoethyl pyrrolinones 261 and 262 to give the yellow oxodipyrromethenes 263 and 264. Catalytic hydrogenation of the benzyloxycarbonyl protecting groups and reduction of the pyrromethenes afforded the aminoethyl oxodipyrromethanes 265a and 266a. Exhaustive methylation followed by brief heating with alkali under argon then gave the desired monovinyloxopyrromethanes 26513 and 266b. These products showed a tendency to disproportionate to neo- and isoneoxanthobilirubic acids 214b and 215b by rearrangement of hydrogen from the methane bridges to the vinyl groups, and hence were used directly in the next stages of the syntheses without purification. The formyl neo- and isoneobilirubic acids 212c and 213c were prepared by thevilsmeier procedure and condensed in aceticacid/hydrobromic acid with the vinyl pyrromethanes 265b and 266b, respectively, to give the
a
CH,NIIZ
CH2NHZ
CH2 Me
p
I
I
H
14
H
0 260
,Me
2
H
261
262
CH2NHZ
I
p N C NHH HN 2 C H 2 N H Z H
0
14 263
Ye
264
Z = C0,CH2Ph
Y
Me I
* p H 26Ja R = CH,CH,NH, 265b R = V
-M e 0
Me
I
H
266a R = CH,CH,NH, 266b R = V
226
The Total Synthesis of Pyrrole Pigments
racemic monovinyl urobilins 256a and 256b (isolated as their hydrochlorides in 6 and 7 % yields, respectively). These synthetic products were difficult to compare173 directly with the natural d-urobilin, but the UV spectrum of 256a showed a band at 230 nm (also occurring in that of the related monocyclic vinylpyrrolinone) which was absent in the UV spectrum of 256b and in that of the natural product. Mass spectral studies were complicated by the tendency of this type of bile pigment to undergo disproportionation in the mass spe~trometer.'~~ Plieninger has also de~cribed"~ syntheses of optically active urobilins of the H I , XIII, and IX series. Optically active neo- and isoneobilirubic acids, 212b and 213b, were prepared by resolution of the corresponding dicarboxylates and decarboxylation either by heating their barium salts in water, or by
E
t
HN
Me
eNil H HN NB
M
N f H aHN t
Me
Me
P
P
e Me
P
P 268
267
Ft
Mc
Et
MC
Me
P 269a
Enantiomer = 269b
warming the quinine salts in acetic acid, a better method which avoids any racemization. The (+)-acids 212b and 213b on self-condensation with orthoformate ester in hydrogen bromidelacetic acid, gave (+)-urobilin-XIII (267) and (+)-urobilin-111 (268), respectively, ( [ a ] : = +4,500 f 100 for both products). In the synthesis of optically active urobilin-IXa the racemic formylisoneobilirubic acid (213c) was condensed separately with both the (+)- and (-)-enantiomen of neobilirubic acid (212b). Each diasterenisomeric
10. Ptodigiosin and Related Compounds
227
mixture produced*7s was crystallized three times to give two urobilin-IX hydrochlorides, the first with [or]: = +4500° and the other with [a]: = -4800O. These were presumably the R R and SS enantiomers (269) and showed pronounced Cotton effects at A1l,,x. 499 nm, positive for the dextrorotatory form, and negative for the levorotatory form. A new and potentially very useful route to pyrrolinones (e.g., 233) has recently been reported,176which makes use of an internal Emmons reaction. Typically a 2,2-diethoxyalkylarnine (270) and an a-diethyl phosphonalkanoic OEt
R
270
27 1
272
H 233
acid (271) in the presence of DCC gave, after acidic work-up, the ketoarnide 272. The latter was cyclized, using sodium hydride, to the required pyrrolinone 233 in an overall yield of about 50%. Applications of this procedure to bile pigment synthesis are awaited with interest.
10. PRODIGIOSIN AND RELATED COMPOUNDS
Prodigiosin is a tripyrrolic red pigment occurring in the bacterium Serrutiu marcescens formerly known as Bacillus prodigiosus. This organism occurs widely in both soil and water, and prodigiosin itself has considerable antibiotic and antifungal activity, but it is too toxic for therapeutic use. A number of possible structures for prodigiosin were put forward by Wrede'?' in 1933, but the one which he later favored178contained a tripyrrylmethene moiety. However, some ten years ago R a p ~ p o rshowed, t ~ ~ ~ by synthesis, that the true structure (273) was a pyrrolylpyrromethene, two of the pyrrole rings being joined together by a direct link. More recently, several analogs have
228
The Total Synthesis of Pyrrole Pigments
been discovered in other bacteria, including nonylprodigiosin180 (274a), undecylprodigiosinlel (274b), metacycloprodigiosin182(275), and a cyclic n ony1prod igiosinre3(276). OMe
C,He
273
J-?f
Ntf
276
275
The crucial piece of evidence in the structure determination of prodigiosin from a mutant was the isolation184of a dipyrrolic compound, C10H10N202, strain of S. marcescens. This compound was shown to be a true precursor of prodigiosin by tracer experiment^;'^' moreover, it could be converted186into prodigiosin by acid-catalyzed condensation with 2-methyl-3-n-amylpyrrole. Rapoport soon afterward showed by that the C,,-precursor was the formyl bipyrrole 277 rather than a dipyrroketone as would have been expected on the tripyrrylmethene formulation. of 277 required the development of new methods for preThe ~ynthesis17~ paring 3-methoxypyrroles and 2,2'-bipyrroles (other methoxy pyrroles were also prepared in the course of this work for comparative purpose^).^^^"^^ OMe
211
Eto2cH/co.Et /
EtO
d,,,,, - g Na
xvlene
/cHzCo2Et HN
I
C0,Et EtO2C
Et0,C
__j
I
C02Et
I
C0,Et
C0,Et
I
278
I ) conc. H2S0,
2) heal
&'
gc0lEt - gc0zE H
C0,Et
219
279
+
OMe
OMe
281
280
-1
dMe C&II
H
273
M~OII/HCI
277
229
230
The Total Synthesis of Pyrrole Pigments
Base-catalyzed condensation of ethyl N-et hoxycarbonylglycinate and diethyl ethoxymethylene malonate afforded the diester 278, which was then methylated, selectively hydrolyzed with concentrated sulfuric acid, and decarboxylated to give the 3-methoxypyrrole-2-carboxylate 279. Previous syntheses of bipyrroles were of little use in the prodigiosin work because they generally led to highly substituted symmetrical products, and drastic conditions were used for Ullman-type coupling reactions. However, an earlier report that A'-pyrroline would couple with pyrrole on heating to give pyrrolidinylpyrrole led Rapoport and H ~ l d e n (after l ~ ~ conducting model experiments) to heat the ester 279 with A'-pyrroline. The resulting pyrrolidinylpyrrole 280 was then dehydrogenated by refluxing in xylene over 5 % palladium on carbon to the bipyrryl ester 281. Conversion of the ester function to an aldehyde was achieved by the McFadyen-Stevens reductionlB8 (because model experiments had revealed that such esters were somewhat inert to aluminum hydrides). The product 277 was found to be identical with the dipyrrolic compound isolated from S. marcescens, and its condensation with 2-methyl-3-amylpyrrole then gave a red pigment identical in all respects to natural prodigiosin (273). Rapoport also investigated other methods for synthesizing bipyrrole~,'~~ because these compounds are clearly of importance not only in relation to prodigiosin but also to the synthesis of vitamin B,, and the corrins.lOOThe most useful outcome of this work was the Vilsmeier-type condensation of 2-pyrrolidone with pyrrole followed by dehydrogenation to give bipyrrole 282 itself. In later work Rapoport showed that the dehydrogenation step could be avoided by the use of 2-pyrrolinones (233);lngb other methods for synthesizing bipyrroles have also since been described by Johnson and Grigg.lBo The synthesis of undecylprodigiosin (274b)was readily accornplished'8' by condensation of the formyl bipyrrole 277 with 2-undecyclpyrrole in ethanolic
Rb H
R'
233
R'
-
282
- ..- ..- . ,?'
.
10. Prodigiosin and Related Compounds
231
hydrogen chloride; this confirmed the structural assignments and also served as a formal total synthesis. Metacycloprodigiosin presents a more formidable synthetic objective owing to the necessity to construct a “meta”-bridged cycloalkylpyrrole (290). However, this was achieved by a multistage procedure starting from cyclododecanone.’82b Sodamide-catalyzed alkylation in glyme gave the 2-ethyl derivative, which was converted into its ketal and brominated with pyridine hydrobromide perbromide. The resulting bromo-ketal (283) was dehydrobrominated with 1,5-diazabicyclo(4,3,0)-non-5-eneto give the unsaturated ketal284, which was hydrolyzed t0 the ethyl cyclodocenone 285. Epoxidation of the double bond gave a mixture of diastereoisomers, which, on treatment with hydrazine in an aqueous ethanol containing a catalytic amount of acetic
’ ~~~~~~~’ -HBr
TosOH 3) PyHErfBr,
0
p+
283
288
284
289
290
275
277
232
The Total Synthesis of Pyrrole Pigments
acid, afforded 4-ethyl-2-cyclododecenol (286). Oxidation of 286 with acid dichromate gave the dodecenone 287, which underwent 1,4-addition of cyanide ion. The cyanoketone 288 was converted into its ketal and reduced to the formyl ketal289 with diisobutyl aluminum hydride. Hydrolysis followed by treatment of the resulting keto-aldehyde with ammonium carbonate then gave the dl-metacyclopyrrole 290. The latter was identical with a pyrolysis product of natural (-)-metacycloprodigiosin and was converted182binto dl-metacycloprodigiosin by condensation with the known formylbipyrrole (277).The product had UV, visible, IR, NMR, and mass spectra identical with those of the natural prodigiosin. The cyclic nonyl prodigiosinle3276 has not yet been synthesized but a number of “unnatural” prodigiosin analogslgl as well as related bipyrroles and p y r r o m e t h e n e ~have ~ ~ ~been synthesized for studies of their chemotherapeutic activity.
11.
CORRINS AND VITAMIN B,,
The corrin ring system 291 is thc parcnt ligand of the structurc of vitamin BIZ
291
(292a)and its coenzyme (292b).Little of the details of the biosynthesis of vitamin B,, is known, but there is no doubt that it is “pyrrole-derived” since i t has been shown that porphobilinogen (36)is a precursor, although the exact point of divergence from the porphyrin pathway is unknown. However, many corrins have been synthesized from pyrrolic interrnediates;lo0 for example, Johnson and ~ o - w o r k e r found s ~ ~ ~ that tetradehydrocorrin salts (293) obtained by cyclization of ac-biladienes can be hydrogenated over Raney nickel at 160” and 100 atm to give a mixture of corrin epimers (294) (obtained as the perchlorates). It is clear that the lack of stereochemical control inherent in this approach makes it of little value for application to vitamin B,, synthesis. It would seem that all routes involving reduction of pyrrolic compounds are equally doomed to failure in view of the presence of no less than nine chiral centers in the vitamin B,, ligand (292).
11. Corrins and Vitamin
H,NOC
\
B,*
233
H2NW 292a
M
I
R = CN
H O OH
1
H,NOC* rv-
Me I
\
NH,
Alternative and more highly controlled syntheses of corrins were therefore required, and in any synthesis of the vitamin itself, it was clearly necessary to pay great attention to stereochemical detail from the outset. Three different approaches to the synthesis of vitamin BI2 are currently under investigation, (a) Eschenmoser’s study of general syntheses of corrinoid compounds, (b) the joint attempt by the Eschenmoser and Woodward groups to develop a specific synthesis of cobyric acid, a vitamin B,, degradation product, and (c) Cornforth’s approach via isoxazole intermediates. Me H
Me Mc
N f ‘N
H 294
ClOh
234
The Total Synthesis of Pyrrole Pigments
A. Eschenmoser’s Approach
The first objective of Eschenmoser and his c o - w o r k e r ~was ~ ~the ~ corrin 295, and their achievement was described in one of the most notable synthetic papers of 1964. Although the target corrin 295 required little attention to stereochemical detail, the route chosen was clearly capable of modification so that it could be used as a basis for the approach to vitamin BIZitself. The chromophore of the corrin macrocycle (291) is nominally made up of three interacting vinylogous amidine systems (296). In view of this fact,
Eschenmoser based his strategy on the concept of carbon-carbon bond formation through condensations involving amide groups (297) activated as their corresponding imino ethers (298) by means of the hitherto little used Meerwein trialkyloxonium salts (see Scheme
!?+h=( 0
OR
+H+
/
297
296 Scheme 10
11. Corrins and Vitamln B,a
235
The best way to accommodate this plan in terms of the “dislocation”196of the molecule was by straightforward “east-west” retrosynthetic division, giving 299 and 300 as the immediate synthetic objectives. Scheme 1 1 shows the synthetic sequence leading to the western (A-D) half (299). Thus Diels-Alder cycloaddition of isoprene and tetramethyl ethylene
H--&
OEt 299
300
tetracarboxylate gave the cyclic tetraester 301, which was converted to the acyclic material 302 by ring cleavage with sodium in liquid ammonia. This was converted to the diamide 303 before creation of the aziridine ring and transformation to the diester 304; the cis configuration was assigned on the basis of mechanistic postulates, the validity of which were later confirmed by degradation of 304 to known substances. The second nitrogen atom in the prospective product 299 was introduced by azidolysis of the aziridine ring, the azide ion attacking at the least hindered carbon atom; the pyrrolidone 305 was obtained by cyclization of this aliphatic intermediate during 15-20 days. The stereochemistry of 305 is predictable from the trans opening of the aziridine ring by the incoming azide ion. Meerwein’s reagent served to activate the lactam function of 305 while still retaining the azide grouping. Condensation with t-butyl cyanoacetate then gave 306, the cis orientation of the amino and ester groups being established on the basis of IR comparisons with model compounds of defined geometry. The azide group in 306 was reduced catalytically, giving the bicyclic monolactam 307, which was transformed to 308 with trifluoroacetic acid and from thence activated to the required A-D component 299 with triethyloxonium tetrafluoroborate. The synthesis of the eastern (B-C) half (300) proceeded along the lines indicated in Scheme 12. The ethyl ester (309) of /?,fl-dimethyllevulinicacid was converted efficiently to @,@-dimethylmethylenebutyrolactam (310) by reaction with hot ethanolic ammonia followed by pyrolysis. Treatment of @,Bdimethyllevulinic acid with thionyl chloride gave the butyrolactone derivative 311, whose sodium salt reacted with the lactam 310 to give the corresponding
236
Me
The Total Synthesis of Pyrrole Pigments
Meo2CXCo2Mc -3
MeO2C
C02Mc
299
J
Scheme 11
N-acylated substance 312; an ingenious photochemically induced migration of the acyl group to carbon in a constant flow system gave 313, which was converted to the bicyclic material 314 on treatment with methanolic ammonia. Pyrolytic or base-promoted dehydration gave 315, which was converted to the sensitive iminoether derivative 300 through the agency of triethyloxonium tetrafluoroborate. The crucial problem of combination of 299 and 300 in the required sense
-
*ie
EtO,C
- a&,
11. Corrins and Vitamin
0
H
309
310
B,*
237
I
311
3+Le
0
Me
Me 314
315
313
300
Me Scheme 12
was solved in the beautiful series of reactions outlined below. The sodium salt of 299 condensed specifically with the iminoether function of 300 to give the open-chain material 316, isolated as its sodium salt; the greater electrophilic reactivity of the conjugated iminoether of 300 compared with the isolated function in 299 meant that none of the unwanted dimer 317 or its further
H--
N
Me
316
The Total Synthesis of Pyrrole Pigments
238
CN
CN
M
e
N f i HN
317
CN
Cloy
Me
318
CN
319
320
reaction products was obtained. Nickel(I1) perchlorate in acetonitrile furnished the chelate 318, largely solving the problem of closure of 316 to acorrin due to the enforced proximity of the appropriate functional groups in rings C and D. The corrin perchlorate (295; X = C104) or hydrochloride (295; X = CI) was obtained in excellent yield merely by treatment of the precorrinoid chelate 318 with potassium r-butoxide followed by the appropriate mineral acid. The cyano group could be removed by hydrolysis and decarboxylation in 0.1M hydrochloric acid at 220" in a greater than 90% yield to give the corrin chloride 319. The structure of the corrin 295 was confirmed in all details by a full three-dimensional X-ray a n a l y s i ~ . ~ ~ * ~ ~ ~ 7 The obvious extension of this synthesis to other metal complexes of corrins has been carried out;1s* in particular the cobalt(II1) chelate (320) has been synthesized and its properties examined. More recent developments*9gfrom Eschenmoser's group have involved the preparation of an A-D moiety by modification of the enamide 310, which had earlier been used only as a precursor for the B-C half (300). Thus this enamide (310) reacted readily with nitromethane in the presence of a catalytic amount of potassium r-butoxide to give the enantiomers 321 which were subjected to Michael addition with methyl acrylate, furnishing the diastereoisomers 322.
11. Corrins and Vitamin Blr
239
After catalytic hydrogenation these gave the corresponding diastereoisomers of the bicyclic material 323, which were separated, and the lower melting
MMe e
P
o
NO2
310
322
M;goE M;g 321
Me
MH" L KCO, Bu'
+
LC0,Me
NH
+
OEt
CN 325
323
324
(undoubtedly anti-) isomer carried through the reaction sequence described below. Treatment with Meerwein's salt gave the bis-iminoether 324, which was condensed with f-butyl cyanoacetate in presence of triethylamine to give 325, the most sterically favored adduct, from which the A-D intermediate 326
.,,ye
Me
EtO &Me Me
CN 326
300
was accessible by treatment with dry trifluoroacetic acid. Condensation with 300 under previously defined conditions*9Bled to the required racemic dicyanocorrin cobalt(II1) complex (328;R = CN) or its hydrolysis and decarboxylation product (328;R = H)(Scheme 13). A major disadvantage of all of the corrin syntheses discussed so far is the essential presence of a metal ion in the product. In most cases this has been
240
The Total Synthesis of Pyrrole Pigments
Me Me
I
CN
Me
327
I
CN
Me Scheme 13
328
cobalt(II1) or nickel(I1) and the vital templating effect of these ions in corrin synthesis has been alluded to. Currently there exists no satisfactory method for removal of such ions to give the metal-free ligand, though such compounds have been an attractive objective, both esthetically and theoretically, for some time. This flaw in existing synthetic strategy has been highlighted even further by reports200 of the isolation and characterization of vitamin B,, compounds lacking cobalt and of the quite remarkable pH dependence of their visible absorption spectra. In addition, certain of their chemical properties are surprising, particularly the relatively drastic conditions required to insert cobalt into the corrin ligand, which had hitherto been regarded as a potent scavenger for various metal ions. Apart from the biosynthetic implications of the natural occurrence of metal-free vitamin BIZcompounds, it was clear that an efficient approach to the free corrin ligand was desirable so that the macrocycle could be studied in detail; the accessibility of such a ligand would also be a good source of many presently unavailable metal complexes. The relatively small quantities of natural materials from photosynthetic bacteriazo0obviously would not suffice for this purpose. The answer to this new problem was supplied by Eschenmoser and Fis~hli,~O~ largely as an extension of the already monumental contributions to
11. Corrins and Vitamin B12
241
corrin chemistry made by the Zurich group. The usual method of templating the ligand into the required geometry was discarded in favor of “sulfide contraction” (Scheme 14). In this method the sulfur atom of a thiolactam (accessible by PzS6treatment of the corresponding lactam), which is sterically unhindered due to the long length of the carbon-sulfur bond, is linked to the methyIidene carbon atom of the enamide (Scheme 14) to give a sulfurbridged intermediate (329). Collapse to the episulfide derivative 330, which
s
0
-
329
330
331
Scheme 14
in the presence of a suitable sulfur acceptor suffers sulfur extrusion201to give the required product 331. This sequence highlights an important principle in synthetic chemistry2O2: Whenever in the synthesis of complex organic molecules one is confronted with a situation where the success of an intermolecular synthetic process is tliivarted by any type of kinetically controlled lack of reactivity, one sliould look for opportunities of altering the structural stage in such a way that the critical synthetic step can proceed intramolecularly rather than intermolecularly. ’’ “
Thus the sodium salt of the precorrinoid ligand 327 was treated with hydrogen sulfide in the presence of trifluoroacetic acid in order to obtain the corresponding ring-A thiolactam derivative. In the event, the compound obtained after “loose” complexation with zinc(I1) was the cyclic isomer 332, which, though unexpected, was adapted for sulfide contraction by treatment with benzoyl peroxide in presence of trifluoroacetic acid giving the ring expanded compound 333 in 72% yield after contact with methanol; the probable mechanistic pathway of this reaction has been outlinedzozand need not be repeated here. Contraction with trifluoroacetic acid in hot dimethylformamide furnished the corrin complexes 334 (R = H)and 334 (R = SH) in about 80% yield, the former already having suffered loss of sulfur. The major product (334; R = SH) was smoothly desulfurized with triphenyl phosphine in the presence of a catalytic amount of trifluoroacetic acid to
242
+M
The Total Synthesis of Pyrrole Pigments
Me
Me
H-- ’N
332
H’ H,
CN
335
N‘
CN
CN
Me
\zAA
333
1
+
t
Me
CN 334
give 334 (R = H)from which the free ligand 335 was liberated with excess trifluoroacetic acid in acetonitrile. The spectroscopic properties of the free corrin have been found to mimic closely those of Toohey’s naturally occurring compound,2ooand a variety of metal complexes of 335 have been obtained in high yield and their properties examined.201 As if to emphasize the deep interest of Eschenmoser’s group in corrins of all types, a totally new route to this macrocycle was devised concurrently with the developments just described, named “The New Road”202*20s and based on a Woodward-Hoffmann type cyclization as the last stage in the synthetic sequence. As a matter of general synthetic principle, it is by far the safest course of action to accommodate any “doubtful” transformations at the earliest possible opportunity, thus providing the maximum amount of latitude for circumventing difficulties. It seems clear in Eschenmoser’s new road that failure of the final cyclization step with its concomitant loss of effort was not considered seriously since the reaction had been deemed “allowed” by the Woodward-Hoffmann rules.204 The ubiquitous enamide 310 was treated with aqueous potassium cyanide giving the derivative 336 with the strongly nucleophilic double bond protected. This was treated with P2S5,furnishing the thiolactam 337 which was oxidized
11. Corrias and Vitamin
B,,
243
with benzoyl peroxide to the reactive disulfide 338; in the presence of the enamide 310 this was smoothly transformed into the bicyclo thio-bridged compound 339. Contraction to the vinylogous amidine 340 was accomplished
336
HN
337
0
Me
Me
CH,
Z
0
M
341
e
Me
-
Me
310
Me
‘ : 2 M . e
340
+-
Me
NC S
0 k
e 339
with triphenylphosphine, and the exocyclic double bond regenerated by treatment with potassium t-butoxide, giving the deprotected intermediate 341. Repetition of this series of reactions with 341 and more of the &sulfide 338 furnished the tricyclic material 342. In this series of operations, the sulfide contraction was catalyzed with boron trifluoride. The silver complex 343 was alkylated with Meerwein’s salt, so avoiding the complications of indiscriminate 0-and N-alkylation which were apparent with the metal free compound 342, and direct treatment with the enamine 344 gave the required open-chain tetracyclic material, isolated as its nickel(I1) chelate (345; M = Ni+). Excess of cyanide ion served to remove the metal from the ligand, laying open a pathway to other chelates [e.g., 345; M = Pd+, Pt+, Co(CN),]. The cyanide-protecting group in such compounds (345) is easily removed with potassium t-butoxide, to give the AID seco-corrinoid systems [346; M = Ni+, Pd+, Co(CN),]; the corresponding complexes (346; M = Zn+, Mg+) are also accessible at this stage from (346; M = Ni+).
244
The Total Synthesis of Pyrrole Pigments
342
344
CN
Me
CN 345
CN 346
The critical cyclization of the seco-corrinoids (346) to corrins appears to depend in an absolutely all-or-nothing sense on the precise nature of the central metal atom. Thus the chelates (346; M = Pd+, Pt+, Zn+, Mg+) all cyclize to the corresponding corrins (347;M = Pd+, Ptf, Zn+, Mg+) in yields greater than 90% by a remarkable, symmetry-allowed, photochemically induced (and formally classified) antarafacial sigmatropic 1,16-hydrogen
11. Corrins and Vitamin BIP
245
transfer and an antarafacial electrocyclic 1,15-.rr,~-isomerization;no cyclization was observed with the cases [346; M = Ni+, Co(CN),]. These observations have prompted a great deal of new theoretical and experimental investigation, and Eschenmoser has already outlined briefly his proposed route to the vitamin BIZdegradation product cobyric acid (348).,02 B.
The Woodward-Eschenmoser Approach
The simplest known natural vitamin B,, derivative205is cobyric acid (348), the synthesis of which would constitute a formal total synthesis of the vitamin since it has been transformed successfully to vitamin B,, by earlier workers. A seemingly trivial, but nonetheless vitally important point of detail is the fact that the 17-propionic side chain must be distinguishable from the remaining side chains so that it may be manipulated specifically in order to exploit Bernhauer’s avenue to the vitamin itself.208 A considerable amount of information has already been published concerning the joint approach by the Woodward and Eschenmoser groups at Harvard and Zurich and has been given in numerous lect~res.~O~ The strategy of Woodward and Eschenmoser’s approach lies in the retrosynthesis of the target cobyric acid (348) to two molecules, the western (A-D) (349) and eastern (B-C) (350) halves. The A-D molecule 349, allocated to the Harvard group, is the more synthetically demanding of the two halves, with its “crowded concatenation of six contiguous asymmetric centers,”207but any Swiss effort spared by this division of labor was to be made up in full by the tremendous wealth of expertise and experience available to the Zurich workers as a result of their several years of successful endeavor in the corrin field. In particular, the method of sulfide contraction, discussed earlier (Scheme 14), which was developed primarily for the coupling of bulkily substituted rings B and C, was to prove invaluable also in the coupling of the A-D and C-B components. Thus the B-C half was divided into the two monocyclic units 351 and 352. The latter, bearing only one chiral center, was obtained from (+)-camphorquinone (353). Ring B (351) was synthesized from 8-methyl-8-acetylacrylic acid (354). the two required chiral centers being specifically created in the required sense by Diels-Alder cycloaddition with butadiene in the presence of stannic chloride; the enantiomeric mixture so produced was resolved by fractional crystallization of the diastereomeric or-phenethylamine salts and eventually furnished the required monocyclic material (355). Chromic acid oxidation served to rupture the double bond and further reaction of one of the newly created carboxyl groups with the ketonic carbonyl group furnished the dilactone acid 356. Homologation of the acetate side chain to propionate
246
The Total Synthesis of Pyrrole Pigments
H,NCO
I
t
t
\
COR
349
\CO,Mc 350
by the Arndt-Eistert procedure gave 357, from which the lactone-lactam 358 was obtained on treatment with ammonia in methanol. Treatment of this material with P,S, furnished the required thiolactam 351, having its acetate side chain concealed as a lactone function. This was to facilitate future activation of the methyl group adjacent to nitrogen as an enamine and the thiolactarn sulfur atom was to be utilized in the sulfur contraction method
3-
0
Me-'
HN
B*
CQMe
H
s
Me
35 1
352
5
0
353
'C0,Me
354
356
357
358
351 247
The Total Synthesis of Pyrrole Pigments
248
s-s
352
0
Mc
S
Me
( ('0,Mc
dm+ 350
0
360
"(
M C
C'O,Me
k0,Me 361
for coupling to ring C (352). Benzoyl peroxide oxidation of 351 in the presence of 352 gave initially the disulfide dimer 359, which collapsed to a B-C sulfur bridged intermediate and was desulfurized directly by heating in triethyl phosphite to give 360 and its epimer 350 in 70% yield; the p-epimer 360 was readily obtained from this mixture in crystalline form. This material has the "wrong" stereochemistry in ring B, but this was not a disadvantage in light of the later discovery that several of the subsequent stages tended to epimerize this center, and furthermore it was known that the 8-position in vitamin B,, can be epimerized under suitable conditions. Experience with the coupling of rings B and C had shown clearly that due to steric factors, it was unlikely that a straightforward iminoether condensation would be satisfactory for joining B-C to D-A and so the lactam in 350
11. Corrins and Vitamin B,,
249
was converted to the corresponding thiolactam derivative (361) as follows. Treatment with methyl mercury isopropoxide served to complex the nitrogen atoms so that reaction with Meerwein’s salt effected 0-alkylation specifically; subsequent decomposition of the intermediate with hydrogen sulfide gave the required activated B-C component (361). The fact that the six contiguous chiral centers in the A-D component 349 exist in the most stable configurations, with large groups trans to each other, provided the ideal opportunity for “stereospecific synthesis by inducti0n,”~07 given the ability to construct the required carbon skeleton. The starting material, 6-methoxydimethylindole (362), though far removed from 349, can be seen to contain certain elements of this component; for example, ring A , in a modified form, can still be discerned, with its two methyl groups, while the benzenoid ring also conceals the carbon skeleton of the ringA propionate side chain, with a carefully sited methoxyl substituent as a “handle” for its later manipulation by Birch reduction. The magnesium derivative of 362 gave the ( f ) indolenine when treated with propargyl bromide. Cyclization with mercuric oxide and boron trifluoride gave the key intermediate, the (&)-tricyclic ketone 364, having the two methyl groups on
Me Me 362
363
364
OCONHMc
NMc
Me fr
365
250
The Total Synthesis of Pyrrole Pigments
the same side of the molecule. The chirality of the second methyl group was dictated by that of the first, since the new five-membered ring must necessarily be cis-fused to the heterocyclic ring. A similar cis fusion must also occur between the two five-membered rings in the alkaloid physostigmine 365, and its congeners, and this has been discussed in detail e l s e ~ h e r e . ~ ~ ~ At this stage the tricyclic ketone was efficiently resolved through the agency of (+)-phenylethylisocyanate, and the absolute stereochemistries of the enantiomers defined by degradative comparison with derivatives of (+)camphor. Experimentally, both of the enantiomers were used, the compounds of the “unnatural” series being utilized ingeniously as models with which to discover transformations to be applied to the “natural” substances* once they were established and optimized. A suitable species (366) from which ring D might be fashioned was next assembled, in an optically pure form by absolute asymmetric synthesis from (-)-camphor as outlined in Scheme 15. (The corresponding enantiomer of366
(-)-Camphor Scheme 15
366
was likewise synthesized from (+)-camphor, for incorporation into the model “unnatural” series.) The acid chloride from 366, when treated with the ring-A precursor 364, gave smoothly the required amide 367, which was induced to cyclize with potassium t-butoxide in t-butanol to the pentacyclic lactam 368. The two new chiral centers created in this cyclization were formed in the required sense by asymmetric induction and the large groups are all in a trans disposition to each other. This complex molecule 368 contains five of the required six contiguous asymmetric carbon atoms, and about 200 grams of this material and its enantiomer were available. The aromatic ring, having served its purpose, was not merely to be discarded, for, as mentioned previously, Birch reduction of this benzenoid ring would eventually lead to the ring-A propionate side chain. Before this could be accomplished, however, it was necessary that certain labile functions within the pentacyclic lactam be protected. The ketonic carbonyl was easily masked
* All of the structures in this report are drawn in the “natural” sense to avoid complication.
11. Corrins and Vitamin B,,
Me 367
Mc
368
251
as its ketal 369 and the lactam group, which was liable to modification under the conditions of the reduction, also needed protection. This was carried out in a more circuitous, but subtly elegant, manner; thus Meerwein’s salt gave the immonium salt 370, which was converted with methoxide in methanol to the orthoamide derivative 371. On heating, 371 furnished the methoxyenamine 372,which withstood perfectly the rigors of Birch reduction with lithium and precisely determined amounts of liquid ammonia, t-butanol, and tetrahydrofuran. The resulting vinyl ether (373)was then readily hydrolyzed to the pentacyclenone 374.Given a Considerable amount of experimental skill this series of reactions could be carried out in remarkably high overall yields approaching 90 % overjue steps! The newly created asymmetric center was, however, entirely in the opposite orientation to that required, although this was not realized until later. The oxygen atom of the spiroketal system of this highly concave molecule lies close to the carbon atom to which a proton must attach itself in the hydrolysis of the vinyl ether 373.As a result of this steric hindrance to protonation the least thermodynamically stable epimer (374)is produced. Since this difficulty was not diagnosed at this point, the synthesis was continued initially without any attempt at rectification of the configuration in ring A. Treatment of the pentacyclenone 374 with further acid liberated the ketone 375,which was transformed via the dioxime to the monooxime 376.Ozonolysis followed by treatment with periodic acid and esterification with diazomethane then gave 377. This series of synthetic operations liberated the propionate side chain from the cyclohexenone ring as well as rupturing the cyclopentene ring to facilitate its conversion to a ring-D species. Quite remarkably, the oxime grouping survived the ozonolysis intact, no doubt due to its highly hindered steric situation; an otherwise unprotected oxime double bond would certainly have been cleaved by ozone. Pyrrolidine acetate-catalyzed cyclization of the diketone system gave 378 by activation of the least hindered carbonyl group through enamine formation. Mesylation
0
--+
H ;
M c II
Me II
Mc
Me
370
369
1
kMc
Mc I((
372
MC
Me
371
Me
Mc Mc 373
252
MC
374
Me
Me
.
375
/
Me
376
CO,Me
I
COCH,
377
MsO,
MsO,
N Me
/t
0
378
H ,' Me ,'
379 253
254
The Total Synthesis of Pyrrole Pigments
and ozonolysis followed by work-up with periodic acid and diazomethane gave 379. This molecule (379) now bore all of the essential features for conversion to a suitable A-D component, the second (ring-D) nitrogen atom of which was ingeniously masked by an oxime function. It was always a major feature of the synthetic strategy that this nitrogen atom should be introduced in this way and then moved into the correct position by means of a Beckmann rearrangement of the oxime. The ketone function of the cyclopentanone ring in the tricyclic ketone 364 was sited in the correct position to serve this purpose and the original ring can still be discerned in the highly complex intermediates at this latestage in the synthesis. Clearly, the strategy of this approach depended on the Beckmann reaction giving the required product, and the fact that failure at this late point could not be tolerated made this perhaps the most well-conceived transformation of the synthesis. Indeed, its experimental execution could not be faulted either, for in an extremely complex series of changes, resulting from treatment of the oxime mesylate 379 in methanol at 170" in the presence of a polystyrene sulfonic acid catalyst for 2 hours, the compound 380 was obtained. Not only had the required rearrangement taken place, but concomitantly with this the diacylamine system was cleaved, and the newly sited nitrogen atom attacked the carbonyl group of the acetyl function, which activated its a-methyl group toward formation of a new sixmembered ring by condensation with its neighboring ester group. The product, given the very apt trivial name "a-corrnorsterone" (380), is on paper the immediate precursor of the required A-D species by hydrolysis of the lactam function (dots in 380 and rotation of ring D about 180"). However, it was found that a vast excess of base was required to efi'ect hydrolysis of the amide function, and it was realized that the propionate side chain of ring A had been created in the wrong orientation. Hence when the lactam ring in acorrnorsterone (380) was opened, the steric compression caused by the presence of two large groupings on the same face of ring A resulted in ring closure; less space was occupied by the groups concerned if the acetate side chain in this ring remained as a lactam grouping. In the Beckmann rearrangement a small amount of a compound isomeric with a-corrnorsterone was also produced, and the ease with which its lactam ring was opened with base showed that this material, /?-corrnorsterone (381), must have had the required trans configuration in ring A , and the anomaly of the Birch reduction was confirmed. This observation, together with the relative rates of hydrolysis of the lactam groups, not only permitted an accurate prediction of the stereochemistries of ring A of the two corrnorsterones (which was subsequently verified by X-ray studies) but also gave a hint of a method by which a-corrnorsterone might be epimerized to its isomer. Clearly, if the lactam ring of a-corrnorsterone could be opened and equilibrated, then due to the
11. Corrins and Vitamin B,,
255
381
380
C0,Me
I
Me
H
‘CH2 C02Me
382
C0,Me C0,Me
tI
“ ‘ O e
‘OMe 383
steric compression experienced by “opened” u-corrnorsterone, the major component of the equilibrium mixture must be the natural b-epimer. Thus heating of u-cormorsterone with a large excess of base followed by acidification and treatment with diazomethane gave /?-corrnorsterone (381) in about 90% yield and this was easily separated from the residual a-epimer. The road therefore seemed open to the A-D intermediate 382, with the future ring-D propionate differentiated from the other esters by its incorporation into a cyclohexanone ring. However, when P-corrnorsterone was opened
256
The Total Synthesis of Pyrrole Pigments
with base and treated with diazomethane in hope of obtaining 382, the amine function of the ring-D vinylogous amide system reacted with the ring-A acetate group to give back /3-corrnorsterone. Experimentally it was found impossible to achieve any synthetically acceptable conditions that would keep the lactam ring open, except for treatment with methanolic hydrogen chloride which gave the corresponding ring-opened compound, “hesperimine” 383, featuring a ring-D vinylogus iminoether group. It was, however, apparent that the nucleophilic character of the vinylic position in hesperimine 383 was considerably less than would be expected from that in the corresponding vinylogous amide (382) were it to be accessible. All the attempts to couple hesperimine 383 with the Zurich B-C component (361) were without reward, and so the 383 was ozonized to the aldehyde 384; the imine function was not affected, no doubt due to the protection afforded to it by the vast array of substituents in its proximity, the aldehyde was then C0,Me
\
C0,Me
\
C0,Me
C0,Me 384
385
C0,Me 386
11.
Me.1
Corrins and Vitamin B,,
257
Me’
1.11
CO~MC
C02Me 387
MeO,C W Me
M
r
C0,Me
e
C0,Me
388
reduced with borohydride to the crystalline alcohol 385. Mesylation with methanesulfonyl chloride in pyridine, followed by treatment with lithium bromide in dimethylformamide, gave the beautifully crystalline bromide 386, the ultimate A-D intermediate. Treatment with the anion of 361 gave the thioiminoether 387, which was the most thermodynamically stable of no less than three such iminoethers. Acid-catalyzed sulfide contraction gave the A-D-C-B material, “corrigenolide” 388, which unfortunately could not be crystallized, possibly as a result of epirnerization of the ring-B and C propionate groups during the coupling process. Phosphorus pentasulfide converted the ring-A lactam to the corresponding thiolactam, at the same time substituting a sulfur atom for the oxygen of the r i n g 8 lactone, to give the dithio derivative 389. Repetition of the series of
258
The Total Synthesis of Pyrrole Pigments
changes established for the synthesis of metal-free corrins was shown by careful scrutiny of electronic absorption spectra to give a mixture of corrinoid compounds; in addition, their mass spectra exhibited the expected molecular ion. However, the relatively low yields experienced in this sequence stimulated experiments to find a more efficient access to the corrin macrocycle from 389 and these, both at Harvard and Zurich, have met with considerable success. Thus treatment of dithiocorrigenolide (389) with dimethylamine led to the labile substance 390, which underwent chelation to 391 with zinc perchlorate. Oxidative coupling with iodine and potassium iodide gave, presumably, the intermediate 392, which, on treatment with trifluoroacetic acid and triphenylphosphine in dirnethylformamide, suffered sulfide contraction and loss of zinc ion; recomplexation with zinc furnished the wellcharacterized zinc chelate 393 in spectroscopic yields between 60 and 70%
MeO,C
wMer C0,Mc
389
C0,Me
C0,Me
390
11.
Corrins and Vitamin B,,
259
C0,Me
?
.H C02Me
Me‘
39 1
C0,Me
)
/CONMe2 C0,Me
C0,Me
‘cO,Me
392
from 389. The zinc in this material could be removed with acid and then replaced with cobalt through the agency of cobalt(l1) chloride in tetrahydrofuran ; cyanide ions and aerial oxidation then furnished the required chelate (394),which could also be prepared in similar yield by the more direct method of base-catalyzed thioiminoether cyclization of the appropriate precorrinoid dicyanocobalt(II1) complex, although no details of this route are as yet available. Meerwein’s salt, followed by aqueous potassium bicarbonate, served to convert the dimethylamide 394 to the “beautifully crystalline”202complex 395 which had been shown to be identical with a sample of authentic material obtainedzoBby removal of the 5- and 15-methyl groups from vitamin Bls by permanganate oxidation.
C0,Me
393
C0,Me
394
CO,Mc
\
CO,Mc
/
C0,Me 395 260
'C0,Me
11. Corrins and Vitamin BIP
261
The remaining problems to be overcome before the vitamin itself can be conquered are only three in number. First, the 5- and 15-methyl groups must be inserted in the macrocycle; considerable work on such reactions with simpler corrinoid compounds has already been reported,210though one might expect the unique steric restrictions of the vitamin itself to make this far from straightforward. Second, the propionate function at the 17-position must be differentiated from all of the other ester groupings. The original strategy had allowed for this, but ozonolysis of hesperimine 383 to the aldehyde 384 was an unexpected deviation, which, though an elegant solution to the immediate problems at the time, destroyed the potentially unique character of the substituent in question. Finally, the complicated problem of mixtures due to epimerization of the ringA, B, and C propionate functions must be solved. C.
Cornforth’s Approach
Although Cornforth has discussed his work many times in lectures,211 which has clearly influenced parts of the Woodward-Eschenmoser synthesis, details have not hitherto appeared in print. His approach to the synthesis of vitamin B,, is based on the established, but previously little appreciated chemistry of the isoxazole ring system. The work was begun in the M.R.C. Laboratories in London in the late 1950s and continued in the Shell laboratories (Milstead) and at Warwick University. lsoxazoles have long been known to undergo catalytic or chemical reduction to P-enamino-ketones, and ring opening by strong base affords p-keto nitriles (see Scheme 16). The ring system is also relatively stable to acid, but
N H,
0
CN
Scheme 16
by virtue of its reductive ring opening, and because isoxazoles are commonly prepared from B-diketones, they may be regarded as a protected form of a b-diketone. The basic strategy of the Cornforth approach involved attempts to construct a macrocyclic intermediate 396 containing three isoxazole units, which it was hoped would undergo reductive fission followed by recyclization to the corrin chromophore (397). The initial objectives, therefore, were to synthesize fragments of the molecule corresponding to the chiral centers of the molecule, in rings A, B, C, and D,each of which had the correct absolute stereochemistry. These were then to be joined by 0-diketone type syntheses, and protected as isoxazoles during the initial construction of a macrocycle of type 396. The first parts to be synthesized corresponded to rings A and B of the macrocycle; an additional advantage of the isoxazole approach lay in
262
The Total Synthesis of Pyrrole Pigments
the fact that they could be synthesized from a common intermediate. The starting material, Hageman’s ester (readily available from acetoacetic ester and formaldehyde) on treatment with cyanide in presence of magnesium acetate underwent I ,4-addition to give a mixture of the epinieric cyanoesters (398).The latter were separated as their semicarbazones and after regeneration with pyruvic acid, the trans-cyanoester, was methanolized to the transketodiester (399). Formylation with methyl formate and base, or better with
Mc
MC
The 396
397
the acetal of dimethyl formamide, afforded the hydroxy methylene (R = CHOMe), or dimethylamino-methylene (R = CHNMe,) derivatives (400). Either of these compounds when treated with hydroxylamine gave the isoxazole (401a) that could be converted by acidic hydrolysis without destroying the isoxazole ring to the corresponding di-acid (401b), which was the initial goal of this part of the synthesis. It was envisaged that alkaline fission at a later stage would result in cleavage of both rings as shown below to generate the acetic and propionic acid side chains of rings A and B (as in 403); alternatively there was the option of opening the isoxazole ring only and converting the resulting keto-nitrile to the base stable, but acid labile, methoxyacrylonitrile (404). However, before continuing with the synthesis, it was essential to check the stereochemistry of the di-acid (401b),and after resolution as the quinine salt the (+)-enantiomer was correlated with a
11. Corrins and Vitamin B,(
263
399
400
4 0 1 ~R = Me 401b R = H
,‘
402 1
II
403,
(Equivalent to ring A or B )
404
steroid degradation product. Subsequently, the di-acid (401b) was converted into a mixture of the monoesters (406a) and (406b), and these were distinguished from each other by a synthesis in which radioactive cyanide was added to the initial Hageman ester; the methyl ester (406b) gave radioactive carbon dioxide on anodic decarboxylation, whereas the other ester (406a) did not. Treatment of the acid chloride derived from 406b with cadmium dimethyl followed by hydrolysis then afforded the keto acid (407). It was later found that 407 could be derived directly and selectively from the intermediate anhydride (405) by treatment with cadmium dimethyl. Transformation of the keto acid (407) to the methylene amide (408) allowed addition of .cyanide or nitromethylene groups as shown; the stereochemistry of the
264
The Total Synthesis of Pyrrole Pigments
401
< H3COCI
Me
405
hk*C J
Mc
406a R 1 = H , R' = Me 406b R' = Me, R' = H
I
i. (C0CI)s i i . Me&d
407
J
i . SO2
ii.
111.
NH, d
CH,NO,
I
410
4 08
I I
HCNIH'
I I
V
H OsC+,,CO\
H
CN
htc Me 409
products 409 and 410 was assumed to be that shown, addition from the top face of 408 being governed by the angular methyl group. The relationship of these two products to their eventual role in the formation of ring A of the corrinoid macrocycle may be discerned from the schematic formulation 411 in which the isoxazole and cyclohexane rings have been assumed to be opened by alkaline hydrolysis. More recently, alternative syntheses of potential ring-A intermediates have been developed at Warwick. Base catalysed addition of nitroethane to Hageman's ester affords a mixture of the epimeric nitro compounds (412);
11.
Corrins and Vitamin B,,
265
these were shown to differ only at the center bearing the nitro group by degradation of both compounds to the known ketodiester (399). The solid isomer 412 (35%) was brominated to give the bromonitroketone (413), and the corresponding ketal was also prepared. Further developments are awaited with interest, but it is envisaged that displacement of bromine from 413 with a suitable carbon nucleophile, followed by elaboration as shown schematically below could give rise to ring-A intermediates.
Rings A and B
Two approaches to the synthesis of ring C were studied in the original work between 1958 and 1961. The first route from a dibromoresorcinol dimethyl ether is illustrated below, a 5-carbon side-chain being introduced into the aromatic nucleus through the lithium aryl and chloroisobutyl methyl ketone. Ketalization of the resulting ketone (414) was effected with ethylene glycol in the presence of p-toluene sulphonic acid together with trimethyl orthoformate, a new procedure at the time, but one which has since become widely used. Lithium-ammonia reduction of the aromatic nucleus followed by oxidative ring opening with periodate then led to the desired ring-C intermediate (415). The second approach to ring C started from camphor quinone (353) and is shown below. It was later adapted for use in the Woodward-Eschenmoser synthesis. Work on the ring-D moiety was also begun in the late 1950s and also seems to have influenced parts of the Woodward-Eschenmoser synthesis. The ct-methyl-a-carbethoxycyclopentanone(416) (available from adipic ester by base catalysed cyclization and alkylation) was reduced to the corresponding alcohol, dehydrated, and then hydrolysed to the acid (417). This was resolved with difficulty as the brucine and quinine salts, and the absolute configuration of the (+)-acid was shown by conversion to the ketone 418 and correlation
Me I
414
1
,.
11.
L~/NII~ 20% ALOII
0ae 0&Me
OH
---+--+-
O
353
CO,Mc
352
266
Me Me
NaIO,
“&Me
Ho2C+COMe CO, H 415
,Mc -
+
Me H*‘ CONH,
HN
I
Me
11. Corrins and Vitamin Blt
267
M e p E i . MefiC02H
417
416
(+I-fenchone
.‘
t
--*
418
of this with (+)-fenchone. Subsequently, the corresponding homologous acid (419) was also prepared and resolved as the dehydroabietylamine. The additional methyl group in this compound (419) was eventually intended to be the C-15 bridge carbon atom of the corrin nucleus, and ring opening by cleavage of the double bond was expected to afford the ring-D propionic acid side-chain. The carboxylic group of this acid was destined to be C-18 of the
I
I
I I
V
CO, H 423 IEaiiivalcnl to ring D )
1
CHsNOI/bate
268
The Total Synthesis of Pyrrole Pigments
corrin nucleus, and various side-chain extension procedures were investigated; these included conversion to the corresponding aldehyde and alcohol, while the alkyl bromide was prepared from 414. However, the most promising approach was chain extension by coupling the acid chloride with malonic ester to give the keto diester (420). Reduction and dehydration then gave the corresponding alkylidene malonate, which underwent addition of nitromethane in presence of base affording the nitromethyl derivative (421). Alternatively, treatment with potassium cyanide afforded the cyanoester (422) with concomitant decarboxylation; other potential intermediates were also synthesized from the alkylidene malonate. Possible routes from either of the two intermediates 421 or 422, to a ring-D intermediate (423) can be discerned by assuming oxidative cleavage of the olefinic double bond as shown above. Many attempts to construct intermediates corresponding to the A-D moiety were also investigated, and one of the more promising approaches is shown below (although this, like the others, failed to give the desired product). Attempts to couple the aldehyde corresponding to the acid 419 with the nitro
419
compound 410 also failed, and clearly, a considerable amount of further work is required in this area before the Cornforth approach can be considered a viable proposition. In addition, the precise manner in which the other intermediates are to be coupled also requires further exploration. Very recently Stevens212*has described some experiments related to the synthesis of corrins, which also utilize isoxazoles; the preliminary reports2’’& include the preparation of model compounds (“seniicorrins”) analogous to the BC portion of the macrocycle involving not only isoxazoles as intermediates, but also y-substituted butyrolactams. The general potential of isoxazoles in organic synthesis has also been recognized by Stork2I3who has shown how they can be used in annelation reactions and in the synthesis of pyridine derivatives.
12.
Postscript
269
12. POSTSCRIPT
Since the original manuscript of this review was written, Stevens has reported his studies on the isoxazole approach to corrinsz*zb, W o o d ~ a r d ~has ’~ described the completion of the joint Harvard-Zurich synthesis of cobyric acid (348), and EschenmoserelShas reported the adaption of his “new road” to a stereochemically controlled synthesis of the same acid. Although the yields in the penultimate stages of the last two investigations are as yet rather low, they constitute total syntheses of vitamin BI2, since the reconversion of cobyric acid to the vitamin has already been described.206At the time of writing the completion of the projects has only been described in lectures, and so only a very brief account can be given. Eschenmoser’s approach to the macrocycle involved the synthesis of an A-D seco-corrinoid following the route he had previously outlined ;eoz rings A, B, and C were synthesized from the lactone (424), while the enantiomer CO Me
<
424
enan”oY
Me0,C-
Me
Me
\ #e 0
H ‘
i I
Me
C0,Me could be converted into ring D; photochemical cyclization of the zinc or cadmium complex of the resulting seco-corrinoid (cf. 346 --t 347) followed by exchange of the metal for cobalt gave a mixture of stereoisomeric cobalt corrins related to 395. Careful control of the reaction conditions ensured that the cyclization gave largely the “natural” series of isomers in which the ring-D acetic ester side chain was trans to the angular methyl group of ring A
270
The Total Synthesis of Pyrrole Pigments
(cf. 348). As in the joint Harvard-Zurich synthesis, a mixture of isomers was produced due to epimerization of the propionic ester side chains of rings A, B, and C. These were separated by high-pressure liquid chromatography, and the two heptamethylesters (395) prepared at Harvard and at Zurich were shown to be identical. The remaining problems to be solved were the insertion of the two mesomethyl groups, differentiation of the 17-propionate function from the other esters, and conversion of the remaining esters to the corresponding amides. Alkylation of a derivative of the macrocycle with benzyloxymethyl chloride followed by brief treatment with thiophenol afforded the 5,1 Sdi(pheny1thiomethyl) derivative; alkylation at the 10-position in this derivative was sterically inhibited by the presence of an additional lactone ring (derived from the 7-acetic acid side chain of ring B) between the 7- and 8-positions of the macrocycle. Raney nickel desulphurization then gave a mixture of products from which the desired meso-dimethyl corrin could be separated by highpressure liquid chromatography. Differentiation of the 7-propionate function from the other propionate and acetate side-chains was achieved by use of a cyanoethyl group introduced before the corrin cyclization. Hydrolysis of the latter with concentrated sulphuric acid then afforded the corresponding amide, together with the ‘neo’-compound due to partial epimerization of the propionate residue in ring C. After separation by high-pressure liquid chromatography, the appropriate amide was shown to be identical in all respects with naturally derived material. The yields at this stage are not yet satisfactory, but it is a tribute to the experimental skill of the investigators and the power of modern spectroscopic methods that fractions of a milligrani of material were separated and crystallized and that the N M R spectrum with less than 100 pg was determined by the Fourier Transform method. Hydrolysis of the amide to the desired 17-propionic acid in presence of the other six ester functions required the development of new techniques, that is, careful nitrosation with N,O, at Harvard and the use of a new reagent at Zurich. Finally, treatment of the hexamethylester mono-acid with ammonia in ethylene glycol gave cobyric acid (395). The completion of this mammoth task, which has taken somc 12 years, represents one of the major successes of modern organic synthesis. In some respects it may be compared with the ascent of Everest or the landing of man on the moon, for the actual achievement of the goal is of little direct value itself except as a symbol of man’s striving for the ultimate in any field of endeavor. Perhaps more important are the new demands these syntheses have made on our chemical skills and ingenuity, to which we might have never otherwise aspired. The spin-off from this work has been immense and will have lasting effects on the future of organic chemistry, for example, the
References
271
myriads of new synthetic methods that have been developed, and the recognition of the potential of earlier discoveries, for example, discoveries of iminoethers by Eschenmoser and of isoxazoles by Cornforth. However, the most profound results of the work arose from a redundant route, followed at Harvard, which led to a consideration of the effects of orbital symmetry on the course and stereochemistry of organic reactions.e1a These have been enshrined in the “Woodward-Hoffmann rules,” and indeed without their aid Eschenmoser could not have contemplated his “new road” to corrins and ultimately to vitamin B,, itself. ACKNOWLEDGEMENTS
We are very grateful to Dr. J. W. Cornforth for providing us with details of his work on the synthesis of vitamin B,, in advance of publication, and we also thank Dr. B. Golding and Professor V. M. Clark for letting us know of the associated work in progress at Warwick. REFERENCES 1. H. Fischer and H. Orth, Die Clretnie des Pyrrols, Vol. 1. Akademische Verlag, Leipzig, 1934. 2. H. Fischer and H. Orth, Die Clreniie des Pyrrols, Vol. 11 (i). Akademische Verlag, Leipzig, 1937. 3. H. Fischer and H. Stern, Die Chemie desPyrrols, Vol. 11 (ii). Akademische Verlag, Leipzig, 1940. 4. T. S. Stevens, in E. H. Rodd, Ed., Chemistry of Carbon Compounds. Elsevier, Amsterdam, 1959, Vol. IVB, p. 1104. 5. R. L. N. Harris, A. W. Johnson, and I. T. Kay, Q. Reu., 20,211 (1966). 6. H. H. Inhoffen, J. W. Buchler, and P.Jager, Fortsrh. Cheiir. Org. Nafrtrsf..26, 284 (1968). 7. G. S. Marks, Heme mid Chlorophyll, Van Nostrand, London, 1969. 8. K. M. Smith, Q. Reu., 25, 31 (1971). 9. 1.U.P.A.C. Rules for Nomenclature, J. Atti. Chern. Soc., 82, 5582 (1960). 10. R. B. Woodward, bid. Cliim. Belge, 21, 1293 (1962). 11. R. Lemberg and J. W. Legge, Haernatin Compounds and Bile Pigrtiettts, Interscience, New York, 1949. 12. A. H. Corwin and E. C. Coolidge, J. Am. Clienr. Sor., 74, 5196 (1952). 13. M. T. Cox, R. Fletcher, A. H. Jackson, G. W. Kenner, and K. M. Smith, Chem. Conrm., 1967, 1141; A. H. Jackson, G. W. Kenner, and K. M. Smith, J . Chem. SOC.(C) 1971, 502. 14. R. L. N. Harris, A. W. Johnson, and I. T. Kay, J. Client. SOC.(C), 1966,22. 15.. A. H. Jackson, G. W.Kenner, G. McGillivray. and G. S. Sach, J. Atti. Client. Soc., 87, 616 (1965).
272
The Total Synthesis of Pyrrole Pigments
16. A. H. Jackson, G. W. Kenner, and D. Warburton, J. Cheni. Soc., 1965,1328. 17. E. Baltazzi and L. 1. Krimen, Client. Reus., 63, 51 1 (1963). 18. P. Rothemund, J . Am. Chern. Soc., 57, 2010 (1935); P. Rothemund and A. R. Menotti, J . Atti. Chetti. Soc., 63, 267 (1941). 19. M.W. Roomi and S. F. MacDonald, Can. J. Chern., 48, 1689 (1970). 20. H.Fischer and E. Fink; Z.p/iysiol Chern., 280, 123 (1944);283, 152 (1948). 21. G.G. Kleinspehn, J. Ani. Cliern. Soc., 77, 1546 (1955);G. G.Kleinspehn and A. H. Corwin, J. Org. Cheni., 25, 1048 (1960). 22. E.Bullock, A. W. Johnson, E. Markham. and K. B. Shaw,J. Cheiti. Soc., 1958,1430; A. W. Johnson and R. Price, Org. Syfi., 42,92 (1962). 23. W.G. Terry, A. H. Jackson, G. W. Kenner, and G. Kornis, J. Clieni. Sac., 1965, 4389. 24. Y. K.Yurev, I. K. Korobitsyna and M. I. Kusnetsova, C/feni.Abs., 45,5680b (1951). 25. B. V. Gregorovitch, K. S. Y. Liang, D. M. Clugston, and S. F. MacDonald, Cari. J . C h i . , 46, 3291 (1968). 26. R. A. Chapman and S. F. MacDonald, see footnote in reference 25. 27. S. F. MacDonald and R. J. Stedman, Cali. J. Cheni., 32,896 (1954). 28. S. F. MacDonald and K. H. Michl, Can. J . Chetn., 34, 1768 (1956). 29. G.P. Arsenault, E. Bullock, and S. F. MacDonald, J. Am. Chem SOC.,82, 4384 (1960). 30. E. J. Tarlton, S. F. MacDonald, and E. Baltazzi,J. Ant. Chefti. SOC.,82,4389 (1960). 31. S. F. MacDonald, J. C/tetft. Soc., 1952,4176. 32. R. G.Westall, Nafrrre, 170, 614 (1952). 33. T. E. Goodwin, Ed., Porphyriris arid Relafed Compoirtids, Biochemical Society Symposium No. 28, Academic Press, London, 1968. 34. S. F. MacDonald and D. M. MacDonald, Can. J . Chefti., 33, 573 (1955); A. H. Jackson, D. M. MacDonald, and S. F. MacDonald, J. A m . Cheni. Soc., 78, 505 (1956); A. H. Jackson and S . F. MacDonald, Cari. J . Chem., 35, 715 (1957); G.P. Arsenault and S. F. MacDonald, Cari. J. C/iem., 39, 2043 (1961). 35. H.Plieninger, P.H a s , and J. Ruppert, C/iem. Ber., 101, 240 (1968). 36. B. Frydman, M. E. Despuy, and H. Rapoport,J. A m Chefti. Soc., 87, 3530 (1965); B. Frydman, S. Reil, M.E. Despuy, and H. Rapoport, 1. Ant. Clietti. Soc.. 91,2338 (1969). 31. A. Albert, He/erocyc/ic Cheniisrry, University of London Press, 1959. 38. G . G. Kleinspehn and A. H. Corwin, J . Atti. Clieni. SOC.,76,5641 (1954). 39. J. A. Ballantinc, A. H.Jackson, G. W. Kenner, and G . McGillivray, Tetrahedron, Suppl. 7 , 241 (I 966). 40. A. F. Mironov, T. R. Ovsepyan, R. P. Evstigneeva, and N. A. Preobrazhenskii, Zh. Obshch. Khini., 35, 324 (1965). 41. H.Fischer, E.Baumann, and H.J. Reidl, Aftit., 475, 237 (1929). 42. A. Hayes, G.W. Kenner, and N. R. Williams, J . Clieni. Soc., 1958, 3779. 43. M. E. Flaugh and H. Rapoport, J. Ant. Chetn. Soc., 90, 6877 (1968). 44. J. M. Osgerby and S. F. MacDonald, Cair. J . C/tetrt., 40, 1585 (1962). 45. P.S. Clezy, A. J. Liepa, A. W. Nichol. and G. A. Smythe, Aiaf. J. Chent., 23, 589 (1970).
References
273
46. H. Fischer and A. Treibs, Ann., 450, 132 (1926). 47. W. Siedel and F. Winkler, Ann., 554, 162 (1943). 48. U. Eisner and R. P. Linstead, J . Chem. Soc.. 1955, 3742. 49. U. Eisner, R. P. Linstead, E. A. Parkes, and E. Stephen, J. Chem. SOC.,1956,1655. 50. U. Eisner, A. Lichtarowicz, and R. P. Linstead, J . Chem. Soc., 1957, 733. 51. A. Treibs and W. 0 1 1 , Ann., 615, 137 (1958). 52. A. W. Johnson, I. T. Kay, E. Markham, R. Price, and K. B. Shaw, J . Clretn. Soc., 1959, 3416. 53. S.Krol, J. Org. Chern., 24,2065 (1959). 54. E. Bullock, A. W. Johnson, E. Markham, and K.B. Shaw, Nafrwe, 185,607 (1960). 55. I. T. Kay, Proc. Nut. Acud. Sci., 48, 901 (1962). 56. A. H. Jackson, P. Johnston, and G. W. Kenner, J. Clienr. Soc., 1964,2262. 57. D. Mauzerall, J. A m Clrenr. SOC., 82,2601 (1960). 58. E. Bullock, Nature, 205, 70 (1965). 59. J. E. Falk, Porplryritrs arid Metalloporphyritrs. Elsevier, Amsterdam, 1964.
60. S. Granick and D. Mauzerall, in D. M. Greenberg, Ed., Metabolic Patlrivays, Vol. 11. Academic Press, New York, 1961, pp. 525-616. 61. L. Bogorad in M. B. Allen, Ed., Comparative Biocheniistry of Photoreactive System, Academic Press, New York 1960, pp. 226-256; J. H.C. Smith, idem pp. 257-277. 62. J. Lascelles, Tetrapyrrole Biosynfhesisnnd Its Regulation. Benjamin, New York, 1964. 63. J. H. Mathewson and A. H. Corwin, J. Am. Chem. Soc., 83, 135 (1961). 64. J. Pluscec and L. Bogorad, Biochern., 9,4736 (1970). 65. (a) H. Fischer and P. Halbig, Ann., 448,193 (1926); (b) H.Fischer and G. Stangler, Ann., 459, 53 (1927). 66. P. A. Burbidge, M.Sc. Thesis, Liverpool, 1963. 67. A. Treibs and H. G. Kolm, Ann. 614, 199 (1958). 68. J. S . Andrews, A. H. Corwin, and A. G. Sharp, J. Ani. Chenr. Soc., 72,491 (1950). 69. J. Wass, Ph.D. Thesis, Liverpool, 1968. 70. R. P. Evstigneeva, V. N. Guryshev, A. F. Mironov, and G . Y . Volodarskaya, Zh. Obshclr. Khitn,, 39, 2558 (1969). 71. G. M. Badger, R. L. N. Harris, and R. A. Jones, Airst. J. Chettr., 17, 987 (1964). 72. R. B. Woodward, W. A. Ayer, J. M. Beaton, F. Bickelhaupt, R. Bonnett, P. Buchsacher, G. L. Closs, H. Duller, J. Hannah, F. P. Hauck, S.110, A. Langemann, E. Le GoK, W. Leimgruber, W.Lwowski, J. Sauer, Z. Valenta. and H. Volz, J. Ant. Chenr. SOC.,82, 3800 (1960). 73. H. Fischer and J. Klarer, Ann., 448, 178 (1926). 74. H. Fischer, H. Friedrich, W. Lamatsch, and K. Morgenroth, Ann., 466, 147 (1928). 75. H.Fischer and A. Kirmann, C o t p f .rend., 189, 467 (1929). 76.
H.Fischer, Natrirwiss.. 17, 611 (1929).
77. See reference 2, pp. 331 and 342.
78. H. Fischer and H. Helberger, Ann.. 480,235 (1930). 79. H. Fischer, H. Berg, and A. Schormuller, A m . , 480, 109 (1930). 82, 4377 (1960). 80. F. Morsingh and S. F. MacDonald, J. Am. Cheni. SOC.,
274
The Total Synthesis of Pyrrole Pigments
81. J. L. Archibald, D. M. Walker, K. B. Shaw, A. Markovac, and S. F. MacDonald, Curt. J . C/teni., 44, 345 (1966). 82. G. S. Marks, D. K. Dougall, E. Bullock, and S. F. MacDonald, J. Ant. Chent. SOC., 82, 3183 (1960). 83. J. W. Purdie and A. S. Holt, Curt. J. Clieni., 43, 3347 (1965). 84. A. S. Holt, D. W. Hughes, H. J. Kende, and J. W. Purdie, J . Ani. Cheni. Soc., 84, 2835 (1962); A. S . Holt, J. W. Purdie, and J. W. F. Wasley, Can. J. Cheni., 44,88 (1966); A. S. Holt in T. W. Goodwin, Ed., The Chemisfry and Biocheniisfry of Plunf Pigntenfs. Academic Books, London, 1965, pp. 3-28. 85. K. M. Smith, Ph.D. Thesis, Liverpool, 1967. 86. J. H. Mathewson, W. R. Richards, and H. Rapoport, Biochem. Biophys. Res. Conrm., 13, 1 (1963); J . Am. Cheni. Soc., 85, 364 (1963). 87. S. F. MacDonald, personal communication (cf. references 25 and 26). 88. P. S. Clezy and A. W. Nichol. Aidsf. J . Chem., 18, 1835 (1965). 89. P. S. Clezy, D. Looney, A. W. Nichol, and G. A. Smythe, Aust. J. Cheni., 19, 1481 (1966). 90. R. Chong, P.S . Clezy, A. J. Liepa, and A. W. Nichol, Ausf. J . C/tent.,22,229 (1969). 91. P. S. Clezy, A. J. Liepa, A. W. Nichol, and G. A. Sniythe, Ausf. J. C/ienr.,23, 589 (1970). 92. P. S. Clezy, A. J. Liepa, and G . A. Smythe, Ausf. J . Chem., 23,603 (1970). 93. P. S. Clezy and A. J. Liepa, Airsf. J . Cheni., 23, 2461 (1970). 94. P. S. Clezy and A. J. Liepa, Aust. J . Chenr., 23, 2477 (1970). 95. R. Bonnett and M. J. Dimsdale, Tetrahedron Lerf., 1968, 731; R. Bonnett, M. J. Dimsdale, and G. F. Stephenson, J. Cheni. SOC.( C ) , 1969, 564. 96. A. H. Jackson, G. W. Kenner, G. McGillivray, and K. M. Smith,J. Chent. SOC. (C), 1968, 294. 97. H. Fischer and A. Schormiiller, Ann., 473, 211 (1929). 98. D. Dolphin, A. W. Johnson, J. Leng, and P. Van den Broek, J. Chon. SOC. (C). 1966, 880. 99. P. Bamfield, R. L. N. Harris, A. W. Johnson, I. T. Kay, and K. W. Shelton, J. Chcnr. Soc. ( C ) . 1966, 1436. 100. A. W. Johnson, Chenr. Brit., 3, 253 (1967). 101.
102. 103. 104. 105. 106.
107. 108.
P. Bamfield, R. Grigg. R. W. Kenyon. and A. W. Johnson, Chem. Conint., 1967, 1029; J . Chenr. Soc. ( C ) , 1968, 1259. R. Grigg, A. W. Johnson, and M. Roche, J . Cheni. SOC. ( C ) , 1970, 1928. A. W. Johnson and I. T. Kay, J . Chem. Soc., 1961,2418; R. Grigg. A. W. Johnson, R. Kenyon. V. B. Math, and K. Richardson, J. Chem. SOC. (C), 1969, 176. G. M. Badger, R. L. N. Harris, and R. A. Jones, Ausf. J. Chem. 17, 1013 (1964). T. Lewis, Postdoctoral Fellowship Report, Liverpool, 1969. A. H. Jackson, G . W. Kenner, and G. S. Sach, J. Chent. Sor. ( C ) , 1967,2045. A. H. Jackson, G. W. Kenner, and J. Wass, unpublished work, R. P. Carr, P. J. Crook, A. H. Jackson, and G. W. Kenner, Chem. Conint., 1967, (C). 1025; R. P. Carr, A. H. Jackson, G. W. Kenner, and G. S. Sach, J. Chent. SOC. 1971, 487.
References
275
109. A. H. Jackson, G. W. Kenner, and J. Wass, Chem. Comm., 1967, 1027. 110. P. J. Crook, Ph.D. Thesis, Liverpool, 1968. P. J. Crook, A. H. Jackson, and G. W. Kenner, J. Chem. SOC.( C ) ,in press. 111. A. H. Jackson. G.W. Kenner, and K. M. Smith, J. Chem. SOC.( C ) , 1968,302. 112. (a) T. T. Howarth, Ph.D. Thesis, Liverpool, 1967; (b) M. T. Cox, Ph.D. Thesis, Liverpool, 1969. 113. M. T.Cox, T. T. Howarth, A. H. Jackson, and G. W. Kenner,J. Am. Chem. Soc., 91, 1232 (1969). 114. R. Fletcher, A. H. Jackson, a n d G . W. Kenner, unpublished work. 115. H. H. Inhoffen, H. Brockmann, K. M. Bliesner, Ann., 730, 173 (1969). 116. R. P. Carr, M. T. Cox, A. H. Jackson, and G. W. Kenner, unpublished work, cf. footnote 1 in reference 113. 117. G. Y. Kennedy, A. H. Jackson, G. W. Kenner, and C. J. Suckling, FEBS Left., 6, 9 (1970); 7, 205 (1970). 118. R. Lemberg, Biocheni J., 29, 1322 (1935); R. Lemberg, B. Cortis-Jones, and M. Norrie, Biochenr. J., 32, 171 (1938). 119. H. Fischer and H . Libowitzsky, Z. Physiol. Chenr., 255,209 (1938); 251,198 (1938); E. Slier, Z. Physiol Chem., 272, 239 (1942); 275, 155 (1942). 120. Cf. R. Lemberg, Reo. Piwe Appl. Chem. (Ausf.),6, 1 (1956). 121. R. Bonnett, M. J. Dimsdale, and K. D. Sales, Chem. Comm., 1970,962. 122. T . Kondo, D. C. Nicholson, A. H. Jackson, and G . W. Kenner, Eiochem. J., 121, 601 (1971); cf. A. H. Jackson and G. W. Kenner in reference 33, p. 1. 123. M. T.Cox, A. H. Jackson, and G. W . Kenner, J. Chem. SOC.(C), 1971, 1974. 124. M. E. Flaugh and H. Rapoport, J . Am. Chem. Soc., 90, 6877 (1968). 125. (a) A. Stoll and E. Wiedemann. Forfschr. Chem. F o r d . , 2, 538 (1952); (b) L. P. Vernon and G. R. Seely, Eds., The Chlorophylls. Academic Press, New York, 1966. 126. H. H. Inhoffen, J. W. Buchler, and R. Thomas, Tefruhedron Left., 1969, 1145. 127. H. Whitlock, R. Hanauer, M. Y. Oester, and B. K. Bower, J. Am. Chem. Soc., 91, 7485 (1969). 128. (a) H. Fischer and G. A. Von Kemnitz, 2.Physiol. Chetn., 96, 309 (1915); (b) A. Treibs and E. Wiedemann, Ann., 471, 146 (1929); (c) reference 3, pp. 144-146; (d) H. Fischer and F. Balaz, Ann., 553, 166 (1942). 129. Cf. R. B. Woodward, Angeiv. Chem., 72,651 (1960). 130. U. Eisner, J . Chem. Soc., 1957, 854. 131. U. Eisner and R. P. Linstead, J. Chem. Soc., 1955, 3749. 132. D. Mauzerall, J. Am. Chem. Soc., 84,2437 (1962). 133. H. H.Inhoffen and P. Jager, Tetrahedron Left., 1964, 1317; 1965, 3387. 134. H. H. Inhoffen, Pure Appl. Chem., 17,443 (1968); see also reference 6. 135. D. Mauzerall, J. Am. Chem. Soc., 82, 2605 (1960). 136. M. Strell, A. Kalojanoff, and H. Koller, Angeiv. Chem,, 72,169 (1960); M. Strell and A. Kalojanoff. Ann., 652, 218 (1962). 137. A. Treibs and R. Schmidt, Ann., 577, 105 (1952); see also reference 128d. 138. M. Strell and A. KalojanotT, Ann., 577.97 (1952); Angeiv. Chem., 66,445 (1954). 139. H. H. Inhoffen, Angeiv. Clrem. h f . , Ed., 3, 322 (1964).
276
The Total Synthesis of Pyrrole Pigments
140. €I. Wolf, H. Brockmann, H. Biere, and H. H. Inhoffen. Aitii.. 704, 208 (1967); I. Fleming, Nature, 216, 151 (1967); J . Chetn. Suc. (C), 1968,2765; H. Brockmann, Airgear. Chent., 80, 233 (1968). 141. H. Brockmann, Atrgew. Cheiii., 80, 234 (1968); H. Brockmann and I. Kleber, Atrgew. Client., 81, 626 (1969). 142. J. J. Katz, G. D. Norman, W. A. Svec, and H. H. Strain, J. Am. Cheiii. Soc., 90, 6841 (1968); H. H. Strain and W. M. Manning, J. Biol. Chefti., 146,275 (1942). 143. R. C. Dougherty, H. H. Strain, W. A. Svec, R. A. Uphaus, and J. J. Katz, J. h i . Chenr. Soc., 88, 5037 (1966); 92, 2826 (1970). 144. W. Rudiger, in reference 33, p. 121. 145. 2.Petryka, D. C. Nicholson, and C. H. Gray, Nafure, 194, 1047 (1962); P. O'Carra and E. Colleran, FEES L e f f . , 5, 295 (1969); R. Bonnet1 and A. F. McDonagh, Cheiii. Cuniiit. 1970, 237. 146. 1'. O'Carra and E. Colleran, Bioclieiii. J . , 115, 13P (1969). 147. J. C. Kendrew. R. E. Dickerson, B. E. Strandberg, R. G. Hart, D. R. Davies, D. C. Phillips, and V. C. Shore, Nufitre, 185, 422 (1960). 148. H. Fischer and A. Treibs, Ann., 457,209 (1927); H. H. lnhoffen, J. H. Fuhrhop and F. v. d. Haar, A m . , 700,92 (1966). 149. H. W. Siegelnian, D. J. Chapman, and W. J. Cole, in reference 33, p. 107. 150. H. W. Siegelman, D. J. Chapman, and W. J. Cole, J. Aiii. Citein. Soc., 89, 3643 (1967); W. Rudiger, P. O'Carra, and C. O'hEocha, Nature, 215, 1477 (1967); H. L. Crespi. U. Smith, and J. J. Katz, Biochent., 7, 2232 (1968). 151. W. Siedel, Z. P/ysiol. Chettt., 245, 257 (1937). 152. H. Fischer and H. Plieninger, Z. Physiol. Cheiti., 274, 231 (1942). 153. H. Fischer and H. Rose, 2. Physiol. Chein., 82,391 (1912); Berichte, 45, 1579 (1912); 0. Piloty and S. J. Thannhauser, Aiiii., 390, 191 (1912). 154. H. Fischer and H. Rose, BericLe, 46, 439 (1913); 0.Piloty and S. J. Thannhauser, Bcrichte, 45, 2393 (1912); 0 . Piloty, Berichre, 46, 1000 (1913). 155. W. Siedel and H. Fischer, Z. Physiol. Chent., 214, 145 (1933). 156. H. Fischer and R. H a s , Z. Physiol. Cheitt., 194, 193 (1930). 157. W. Siedel, Z. Pltysiol. C h i . , 237, 8 (1935). 158. H. Fischer and E. Adler, Z. Physiol. Cheni., 200, 209 (1931). 159. H. Fischer and E. Adler, Z. Physiol. Chein., 212, 146 (1932). 160. Reference 2, p. 621 ef seq. 161. W. Siedel, 2. Physiol. Cheiti., 231, 167 (1935). 162. H. Fischer and H. Hofelniann, Z. Physiol. Cheiit., 251, 187 (1937). 163. W. Siedel and E. Meier, Z. Physiol. Cheiii., 242, 101 (1936). 164. W. Siedel and H. Moiler, Z. Physiol. Cheitt., 264, 64 (1940). 165. J. H. Atkinson, R. S. Atkinson, and A. W. Johnson, J. Chent. SOC.,1964, 5999. 166. D. R. Ridyard, M.Sc. Thesis, Liverpool, 1966. 167. H. Plieninger and M. Decker, Aim., 598, 198 (1956); H. Plieninger and J. Kurze, Aim., 680, 60 (1964). 168. H. Plieninger, Lecture to SocietC Vaudoise des Sciences Naturelles, Lausanne, October, 1963, abstracted in Chiiitiu, 17, 393 (1963); H. Plieninger and U. Lerch, Aim., 698, 196 (1966).
References 169. 170. 171. 172. 173. 174.
175. 176. 177.
277
H. Plieninger and J. Ruppert. Ann., 736, 43 (1970). C. H. Gray and D. C. Nicholson, J . Cherri. Soc., 1958, 3085. W. Siedel and E. Meier, Z. Physiol. Chenr., 242, 121 (1936). C. J. Watson, A. Moscowitz, D. A. Lightner, Z. J. Petryka, E. Davis, and M. Weiner, Proc. Nut. Acad. Sci. US.,58, 1957 (1967). H. Plieninger and R. Steinstrasser, Ann., 723, 149 (1969). A. H. Jackson, K. M. Smith, C. H. Gray, and D. C. Nicholson, N a m e , 209, 581 (1966); A. H. Jackson, G. W. Kenner, H. Budzikiewicz, C. Djerassi, and J. M. Wilson, Telrahedron,23, 603 (1967). H. Plieninger, K. Ehl, and A. Tapia, Ann., 736, 62 (1970). G. Stork and R. Matthews, Chenr. Comm., 1970,445. F. Wrede and A. Rothaas, Z. Physiol. Chem., 215, 67 (1933); 219, 267 (1933); 222,
203 (1933); 226,95 (1934). 178. H. Fischer and K. Gangl, 2. Physiol. Chern., 267, 201 (1941); A. Treibs and K. Hintermeier. Ann., 605, 35 (1957); A. J. Castro, A. H. Corwin, J. F. Deck, and P. E. Wei, J. Org. Chem., 24, 1437 (1959). 84,635 (1962). 179. H. Rapoport and K.G. Holden, J . Am. Chem. SOC., N. N. Gerber, Appl. Microbiol., 18, 1 (1969). 180. 181. H. H. Wasserman, G. C. Rodger, and D. D. Keith, Chenr. Comm., 1966, 825; K. Harashima, N. Tsuchida, and J. Nagatsu, Agric. Eiol. Chem. (Japan), 30, 309 (1966); K. Harashima, N. Tsuchida, T. Tanaka, and J. Nagatsu, Agric. Biol. Chem. (Japan), 31,481 (1967). 182. (a) H.H. Wasserman, G. C. Rodgers, and D. D. Keith, J . Am. Chern. Soc., 91, 1263 (1969); (b) H. H. Wasserman, D. D. Keith, and J. Nadelson,J. Am. Clienr.SOC., 91, 1264 (1969). 183. N. N. Gerber, Tetrahedrorr Lerr., 1970, 809. 184. U. V. Santer and H. J. Vogel, Fed. Proc., 15, 345 (1956); Bioclrerti. Biophys. Acta, 19, 578 (1956). 185. H. H. Wasserman, J. E. McKeon. L. Smith, and P. Forgione, J . Am. Chern. SOC.,82, 506 (1960). 186. H. Rapoport and C. D. Willson, J. Am. C/rem. SOC.,84, 630 (1962). 80, 6249 (1958). 187. D. W.Fuhlhage and C. A. Vanderwerf, J. Am. Clrem. SOC., 188. E. Mosettig, in Organic Reactions, Vol. VIII. Wiley, New York, 1954, p. 218; R. G. Jones and K. C.McLaughlin, J. Am. Clrenr. Sac., 71, 2444 (1949). 189. (a) H. Rapoport and N. Castagnoli, J. Am. Clrern. Soc., 84, 2178 (1962); (b); J. Bordner and H. Rapoport, J. Org. Clienr., 30,3824 (1965). 1963,359; E. Bullock, 190. R. Grigg, A. W. Johnson, and J. W. F. Wasley,J. Chem. SOC., R. Grigg, A. W. Johnson, and J. W. F. Wasley, J. Clieni. Soc., 1963,2326; R. Grigg and A. W. Johnson, J. Chem. SOC.,1964,3315. 191. W. R. Hearn, M. K.Elson, R. H. Williams, and J. Medina-Castro, J. Org. Cliem., 35, 142 (1970). 192. A. Ermili and A. J. Castro, J. Hererocyc. Chenr., 3,521 (1966); A. J. Castro, G. R. Gale, G. E. Means, and E. Tertzakian, J. Med. Chem., 10,29 (1967). 193. R. Grigg, A. W. Johnson, and P. v. d. Broek, Chem. Comm., 1967,502; I. D. Dicker, R. Grigg, A. W. Johnson, H. Pinnock, K. Richardson, and P. v. d. Broek,J. Chem. SOC.(C), 1971,536.
278
The Total Synthesis of Pyrrole Pigments
F. Elsinger, A. Eschenmoser, I. Felner, H. P. Gribi, H. Gschwend, E. F. Meyer, M. Pesaro, and R. Scheffold, Angew. Chem., 76,
194. E. Bertcle, H. Boos, J. D. Dunitz,
393 (1964). 195. H. Meerwein, H. Hinz, P. Hoffmann, E. Kroning, and E. Pfeil, J. Prakt. Chem., 147, 17 (1937). 196. Cf. E. J. Corey, Pure Appl. Chem., 14, 19 (1967). 197. J. D. Dunitz and E. F. Meyer, Proc. Roy. Soc. (A), 288,324 (1965). 198. A. Eschenmoser, R. Scheffold, E. Bertele, M. Pesaro, and H.Gschwend, Proc. Roy. Sor. ( A ) , 288, 306 (1965). 199. 1. Felner, A. Fischli, A. Wick, M. Pesaro, D. Bormann, E. L. Winnacker, and A. Eschenmoser, Angeiv. Chem., 79, 863 (1967). 200. J. 1. Toohey, Proc. N a f . Acad. Sci. US.,54,934 (1965); Fed. Proc., 25,1628 (1966); K. Sato, S. Shimzu, and S. Fukui, Biochem. Biophys. Res. Comm., 39, 170 (1970). 201. A. Fischli and A. Eschenmoser, Angew. Chem., 79, 865 (1967). 202. A. Eschenmoser, Q. Rev.. 24,366 (1970). 203. Y. Yamada, D. Miljkovic, P. Wehrli, B. Golding, P. Loeliger, R. Keese, K.Mtiller, and A. Eschenmoser, AtKeiv. Chem., 81, 301 (1969). 204. R. B. Woodward and R. Hoffmann, Angea. Chem., 81,797 (1969). 205. R. Bonnelt, Chetn. Revs., 63, 573 (1963). 206. W. Friedrich, G. Gross, K. Bernhauer, and P. Zeller, Helu. Chin?.Acta, 43, 704 ( 1960). 207. R. B. Woodward, Piire Appl. Chem., 17, 519 (1968). 208. A. H. Jackson, Ph.D. Thesis, Cambridge, 1954; referred to by B. Robinson, Chem. Ind., 1963,218; and by E. Coxworth in The Alkaloids, Vol. 8, R. H. F. Manske, Ed. Academic Press, New York. 1965, p. 27. 209. K. Bernhauer and F. Wagncr, unpublishcd work. 210. D. Borman, A. Fischli, R. Kcese, and A. Eschenmoser, Angeiv. Chem., 79, 867 (1 967). 21 I . Cf. J. W. Cornforth (reported by P. B. D. de la Mare), Nature, 195,441 (1962). 212. (a) R . V. Stevens, L. E. DuPree, Jr.. and M. P. Wentland, Chem. Comrn., 1970, 821; R . V. Stevens and M. Kaplan, idem, 1970, 822; (b) R. V. Stevens, C. G. Christensen, W. L. Edmonson, M. Kaplan, E. B. Reid and M. P. Wentland; J. Amer. Chem. SOC.,93, 6629 (1971); R. V. Stevens, L. E. DuPree Jr., W. L. Edmonson, L. L. Magid and M. P. Westland, J. Anier. Chem. SOC.,93, 6637 (1971). 213. G. Stork, S. Danishefsky, and M. Ohashi; J. Am. Chem. SOC.89, 5459 (1967); M. Ohashi, H. Kamachi, H. Kakisawa, and G. Stork, idem, 89, 5460 (1967); G. Stork and J. E. McMurry, idem, 89,5463 and 5464 (1967). 214. R. B. Woodward, XXIIIrd International Congress of Piire and Applied Chemistry, Boston, July 1971. Vol. 2, Butterworths, London, 1972; IUPAC Symposium on Nafiiral Prodiicts, Delhi, February 1972. 215. A. Eschenmoser, "W. Baker Lecture"; Bristol. May 1972; cf. XXIIIrd International Congress of Piite and Applied Chemistry, Boston, July 1971, Vol. 2, Butterworths, London, 1972. 216. R. B. Woodward and R. Hoffmann, The Conservation of Orbital Symmetry, Verlag Chemie, Weinheim, 1970.
Total Synthesis OfNatural Products, Volume1 Edited by John Apsimon Copyright © 1973 by John Wiley & Sons, Inc.
The Total Synthesis of Nucleic Acids* SARAN A. NARANG Biochemisrry Laborarory, National Research Councilof Canada, Otrawa. Canada AND
ROBERT H. WIGHTMAN Deparrmenr of Chemistry, Carleton Universiry, Otlawa, Ontario, Canada
1. Introduction 2. The Generalized Problem 3. Phosphorylating and Condensing Reagents 4. Protecting Groups A. Primary Amino Function B. For 5‘-Hydroxyl Function (Primary) C. For 3’-Hydroxyl Function (Secondary) D. In the Ribonucleotides E. For Phosphate 5. Chemical Syntheses of Polynucleotides A. Polymerization Method B. Stepwise Condensation Methods for Polynucleotides without a Terminal Phosphate C. Stepwise Synthesis of Deoxyribopolynucleotides Bearing a 5’-Phosphomonoester End Group D. Polymer-Support Synthesis
280 282 284 286 286 288 288 290 29 I 293 293
300 305 309 279
280
The Tofal Synthesis of Nucleic Acids
E. Synthesis of Oligoribonucleorides 6. Enzymes as Rcagcnts A. DNA-Polymerase B. Nucleotidyl Transferase Enzyme C. RNA-Polymerase (DNA-Dependent) D. I’olynucleotide Phosphorylasc E. Ribonuclease Enzymes F. I’olynucleotide Kinase G . Alkaline Phosphatase ( E . coli) H . Phosphodiesterases (Spleen and Venom) 1. Polynucleotide Ligase 7. Gene Synthesis 8. Conclusion References
1.
311 313 314 316 316 317 318 319 319 319 320 32 I 324 325
INTRODUCTION
The hypothesis that the genes in all living organisms contain all the information required for the cell to reproduce is now more than 50 years old. I n chemical terms the basic idea of this hypothesis has been that the deoxyribonucleic acid (DNA) in all living organisms directs the synthesis of all the protein required by that organism. This control is exerted through the intermediacy of ribonucleic acids (messenger or mRNA, transfer or tRNA), which transcribe and transmit the information originally present in DNA. Although most of the predictions of the “central dogma” have been verified using biological approaches, the most successful and direct confirmation has been obtained by the use of synthetic DNA molecules.’ For many such investigations supplies of these molecules would be essential, but their pure form may not be derived from degradative procedures. Thus there have been attempts at a rational synthesis of such molecules not only because of the challenges of their chemical complexity but also for their usefulness in studying some cardinal biological processes. Naturally occurring DNA consists of two long complementary polynucleotide chains twisted about each other in a regular helix. Each chain contains a large number of nucleosides joined through a C,. + C,, internucleotide phosphodiester bond. The two chains are held together by hydrogen bonds between pairs of bases in which guanine (purine) is always joined to cytosine (pyrimidine) and adenine (purine) always bonds to thymine (pyrimidine) or to uridine (Pyrimidine) in RNA,* as illustrated in Fig. I . Individual DNA molecules may be very large; in fact, the E. coli chromosome * NRCC No. I I844
1. Introduction
281
H Figure 1. Base pairing in DNA due to hydrogen-bonding (note that in RNA thymine is replaced by uridinc).
is probably a single DNA molecule with M W -2-4 x loo. Most DNA molecules correspond not to single genes but to a collection of genes. The average molecular weight is about los (Le., 1500 nucleotide pairs). Much preliminary work on the individual nucleosides (heterocyclic base plus sugar), nucleotides (base, sugar and phosphate), and small oligonucleotides (short chains) has shown that it is not feasible to attempt a purely
* The basic system of abbrcviations uscd for polynuclcotidcs and their protected derivatives is as used in J. B i d . Chern. Thus the single letters A, T, C, G, and U represent the niicleosides of respectively adenine, thymine, cytosine, guanine, and uridine. The letter p to the left of the nucleoside initial indicates a 5’-phosphomonwster group and the same letter to the right indicates a 3‘-phosphornonoester group. Hence, in going from the left to the right the polynucleotide chain is specified in the C,, -+ C,, direction. The protecting groups on the purine or pyrimidine rings are designated by two-letter abbreviations added as superscripts after the nucleoside initial: thus ADZfor N-benzoyldcoxyadenosine, CAnfor Nanisoyldeoxycytidine, GAc for N-acetyldeoxyguanosine. The acetyl group at the 3’hydroxyl group of a nucleoside is shown by -0Ac added after the nucleoside initial: thus d-pTpGAc-OAc for the dinucleotide, 5’-O-phosphorylthymidylyl-(3’4 5’)-N, 3‘-0diacetyldeoxyguanosine.
282
The Total Synthesis of Nucleic Acids
chemical synthesis of such a large molecule. The first concerted efforts to synthesize some of the simpler compounds were due to Todd and his coworkers.2 Although these synthetic studies were primarily intended as final structure proofs, they did succeed in emphasizing some of the critical problems. Nucleotides and their polymers are sensitive molecules and because of their polar nature they are not amenable to the usual manipulations of organic chemistry. Also, methods for introduction of a phosphoric acid residue and specific formation of the internucleotide bond were relatively crude. The modern phase of nucleic acid chemistry and synthesis rests on the development of appropriate condensing reagents, the application of ionexchange chromatography for separation and purification of products, and the use of various enzymes as specific reagents for the characterization and synthesis of potynucleotides. The 1960s have witnessed tremendous strides in solving the problems associated with the synthesis of small chains of deoxyribo- (and ribo-) oligonucleotides. As a result much valuable information has been obtained concerning the genetic code, especially through the efforts of Khorana and his c o - ~ o r k e r s .However, ~ there seeins little doubt that new concepts and approaches have to be explored to achieve more efficient and easier syntheses of the longer chains. A few comprehensive articles have already been written concerning chemistry and synthesis in the nucleic acid field. These include an excellent book4 and a small m ~ n o g r a p h ,both ~ somewhat out-of-date; a reference texta pertaining primarily to manipulations of individual nucleosides and nucleotides; and some shorter reviews.’ In addition much recent work has been reported on synthetic modificationse of the naturally occurring molecules. Some of these exhibit promising physiological a c t i ~ i t y Further, .~ the very interesting “rare” bases present in tRNA are beginning to receive at tent ion. l o With this review, which deals only with the development of procedures for the synthesis of deoxyribo- (and ribo-) oligonucleotides during the period 1960-1970, we hope to achieve two objectives. First, we attempt to mention all pertinent investigations and emphasize those approaches that have been most successful in practice. Second, we hope to illustrate some of the areas where improvements would prove most beneficial. 2. THE GENERALIZED PROBLEM
To achieve any success in nucleic acid synthesis one should be familiar with the organic chemistry of carbohydrates, nitrogen heterocycles, and phosphates esters as well as the handling of an increasing number of pertinent
Y
I
/NH
xo-cQ'
7-
0-P-0-
OH
+
I
/
1) remove X
2) condense wilh phosphorylatinp agent
WO
3) remove all proleclino proups
I
J
INH
WP
Y
I I
IH
(W)O-P=O
OH -
o-c@y
OH
0-
+
I
o=p-oI
Figure 2. Schematic representation of possible approaches to a deoxydinuclcotide diphosphate (B = heterocyclic base; W, X, Y. and 2 = protecting groups for phosphate, primary hydroxyl, primary amino, and secondary hydroxyl respectively). 283
284
The Tolal Synthesis of Nucleic Acids
enzymes. The fundamental objective of the problem is efficient formation of the internucleotidic phosphodiester bond specifically between the CB’and C,. positions of two adjacent nucleosides. This is the natural bond occurring in D N A and R N A and is usually formed by reaction between a free phosphate group and a free hydroxyl group situated in the correct positions of the appropriate nucleosides or nucleotides. In order to perform this union most selectively it is often of paramount importance to protect any other functional groups, which may include (a) primary amino groups on the heterocyclic ring, (b) 5’-hydroxyl (primary), (c) 3’-hydroxyl (secondary) and/or 2‘-hydroxyl in the ribo series, (d) phosphate groups at the 3‘- or 5’-positions. Figure 2 illustrates two generalized approaches. Thus a major portion of the synthetic work has been devoted to the fol lowing : 1. The improvement of methods for activating the phosphate ester group in a nucleotide so as to effect phosphorylation of the hydroxyl on another nucleotide or nucleoside. 2. The design of suitable protecting groups for the various functionalities; they must be quantitatively introduced and readily but selectively removed. 3. The development of. effective techniques for the purification and characterization of the resulting polynucleotides, since the condensation reactions currently in use are neither quantitative nor free of contaminating side products. 4. The elaboration of methods for the polymerization of mono- or block units. 5 . The utilization of certain enzymes as specific reagents in the “multiplication-amplification” of polynucleotide templates or the linking of chains for the synthesis of longer oligonucletoides containing hetero sequences. 3. PHOSPHORYLATING AND CONDENSING REAGENTS
Although most simple nucleotides are now readily available from commercial sources it is occasionally necessary to introduce phosphate (sometimes containing P32)into the molecule. Two reviews are available on phosphorylating agents“ and several examples follow in later sections of this essay. Most commonly, two of the phosphate oxygens are blocked and this reagent is then activated, either directly (e.g., phosphorochloridates, 1 in Fig. 3) or indirectly via the condensing agent, for attack by an appropriate hydroxyl group from the nucleoside. Subsequent removal of the protecting group generates the free nucleotide which may then undergo a further reaction. Some of the more recent phosphorylating reagents include pyrophosphoryl chloride,’*“ 0-phenylene phosphorochloridate,12band trimetaphosphate.12c
3.
Phosphoryiatlng and Condensing Reagents
0
PhO
/-\
PhO
CI
0
0 Nucleoside-0-P-X
II
II
Nucleoside-0-P-0-Y
I
I
02a
0
F = IN
X=N
2b X = F
4
285
+
3a
Y
0-
It
= -C-OEt
5
6
Figure 3. Condensing agents investigated for nucleic acid synthesis.
Condensation, or formation of the internucleotidic bond, is usually accomplished by reaction of a free hydroxyl function of one nucleoside or nucleotide with the reactive phosphate of another nucleotide. Various approaches have been reported for making the phosphate amenable to nucleophilic attack by hydroxyl. Some of these are illustrated in Fig. 3 and include: 1. Formation of derivatives such as imidazolide (ZII)'~~ and phosphorofluoridate (Zb).I4 2. Formation of mixed anhydrides with acid chlorides, for example, ethyl chloroformate (3a),16 mesitylene (or triisopropylbenzene) sulfonyl chloride (3b),16 picryl chloride (3c).17
286
The Total Synthesis of Nucleic Acids
3. Reaction with dehydrating agents such as dicyclohexylcarbodiimide
(4).'* 4. Reaction with miscellaneous activating agents, for example, isoxazolium trichloroacetonitrile.*l salts (5),l9dimethylformamide chloride
A comparative study of some of these condensing reagents was reported in 196422 and since that time dicyclohexylcarbodiimide (DCC) and mesitylenesulfonyl chloride (MS)or triisopropylbenzenesulfonyl chloride (TPS) have been used almost exclusively. Since the hydroxyl group should be the poorest nucleophile available, all other functional groups are usually blocked and an excess of the phosphate-containing moiety is used when feasible. Under the usual experimental conditions, both reagents produce undesirable side products that often require extensive chromatographic separations while the DCC reaction is slow (2-3 days). Additionally, unless a large excess of the phosphate-containing component is used, there is a dramatic decrease in yield as the nucleotide chain grows larger than 8-10 units. Clearly, this is an area of nucleic acid chemistry that still requires considerable effort and ingenuity. For example, a recent report of new reagents for peptide synthesis23has promised to investigate their applicability for nucleic acids.
4. PROTECTING GROUPS
A.
Primary Amino Function
Several of the heterocyclic bases (A, C, G) contain reactive primary amino functions which apparently must be blocked, preferably for an entire synthesis. * N-Acylation (acetyl, benzoyl, anisoyl) is most frequently employed as a method of protection. The general approach has been to fully acylate the mononucleotide or nucleoside and then selectively liberate the hydroxyl groups by taking advantage of the facile base hydrolysis of esters compared to amides. Commonly encountered conditions involve a brief treatment with IN aqueous sodium hydroxide at room temperature or below. Additionally, aromatic amides are known to be more stable than aliphatic amides at high pH, apparently due to the ionization depicted in (7)(Fig. 4). This observation has led to a marked preference for the benzoyl or anisoyl groups, and excellent yields of the protected products, for example Nbenzoyl deoxyadenosine (8), have been obtained24 by the sequence outlined in Fig. 4.
* See, for instance, reference 5 (p. 105), reference 15 or reference 12 in P.T. Gilham and H. G. Khorana, J. Am. Chern. SOC.,81, 4647 (1959). However, for reactions without amino-protection, see A. M. Michelson, J. Cliem. Soc., 1959, 1371 and reference 30.
Sugar 7
deoxyadenosine 5'-phosphate
-*C,H&OCI pyrldine
V 6 ,
CeH6
0
OH-
00
I
,c=o
CeH,
'c-0 I NH
0
I
0OH 8
288
The Total Synthesis of Nucleic Acids
As well as acetyl, another nonaromatic protecting group recently introd ~ c e d ,involved ~~ treatment with isobutyloxy chloroformate. After the sequence just outlined, good yields of the protected products, for example, N-isobutyloxycarbonyl deoxycytidine 5’-phosphate (9) (abbreviated pCBoC), could be obtained. All of these groups can be removed readily by treatment with concentrated ammonium hydroxide for 2 hours at 50°C or 2 days at room temperature. A selective protection of the amino group can also be accomplished by treatment with dimethylformamide dimethylacetal or the diethylacetal. Some complications involving transacetalization of the cis-2’,3’-diol system in ribonucleosides can be overcome by aqueous hydrolysis. Thus, N-dimethylaminomethylene derivatives of all the required nucleosides, such as deoxyguanosine (lo), have been prepared and used in oligonucleotide syntheses.26 In contrast to the acyl derivatives, this protecting group can be removed easily in either weakly alkaline or weakly acidic media. B. For 5’-Hydroxyl Function (Primary)
Triphenylmethyl or trityl (lla), the standard group for the selective protection of a primary hydroxyl function, was found to be suitable for any synthesis involving only pyrimidines2’ However, the conditions required for its removal (80% aqueous acetic acid at 100°C) caused extensive decomposition in the purine nucleotides due to the acid lability of the glycosyl bond. Therefore, more sensitive groups such as the mono-llb or dimethoxyl l c derivatives were developed.28These can be easily removed by treatment with 80% acetic acid at room temperature. A bulky, base-labile group was reported earlier for t h y ~ n i d i n eand ~ ~ has recently been reintroduced for the ribonucle~tides.~~ Treatment of the nucleosides with pivaloyl chloride in pyridine causes selective protection of the 5’-hydroxyl (see 12). This group can be readily removed by treatment with methanolic tetraethylammonium hydroxide for 2-3 hours at room temperature. Some alkoxy-substituted acetyl derivatives have also been tested on nucleosides in an attempt to achieve selectivity at the 5’-hydroxyL31 C. For 3’-Hydroxyl Function (Secondary)
By far the most common method of protecting this hydroxyl function is simple acylation; acetylation (see 13a) is most frequently encountered, followed by b e n ~ o y l a t i o n The . ~ ~ acetyl group which is readily introduced using acetic anhydride or acetyl chloride can be removed under very mild conditions (2N aqueous sodium hydroxide at 0°C for 5-10 minutes) and thus allows a high degree of selectivitya33
4.
Protecting Groups
289
0
HO 11s R = R’ = H l l b R = H; R’ = -OCH3 l l c R = R’ = -OCH3
OH
12
0
0OR 13a R = CH3CO-
I
13b R = O=C-CH&Hz-CPh
0
I1
Figure 5. Protection of 5’-(primary) and 3’-(secondary) hydroxyl.
Recently, the fl-benzoylpropionyl group (13b) has been introduced as a protecting group for this position.34It can be cleaved quantitatively within a few hours at room temperature by using dilute solutions of hydrazine hydrate in pyridine/acetic acid. However, the N-benzoyl groups of deoxycytidine and deoxyadenosine are also labile under these conditions. A dinitrophenylsulfenyI group (13b) susceptible to cleavage by thiophenol or Raney nickel was r n e n t i ~ n e d . ~ ~ In an attempt to achieve more selective removal, an increasing variety of other reagents suitable for protection of this functionality have been reported. Generally, they have been tested only on nucleosides and not thus
290
The Total Synthesis of Nucleic Acids
far, employed in any extended synthesis. As a result, derivatives from reaction with P,P,fi-tribromoethyl chloroformate have been cleaved using a zinccopper couple,36 while dihydrothiophene adducts are labile to silver Reaction with 2-chloroethyl orthoformate produces products which can be deblocked with 80 % acetic acid at room temperature,38 whereas chloroacetic anhydride39and p-nitro-chloroformateao provide derivatives which are most readily cleaved by organic amines at room temperature. F ~ r m y land ~~ benzoyl f ~ r m y l ~nucleosides ~'' have also been mentioned as base-sensitive compounds. Esters of dihydrocinnamic acid can be specifically removed by the enzyme c h y m o t r y p ~ i n . ~ ~
D. In the Ribonucleotides An additional problem in ribooligonucleotide synthesis hinges on the protection of the 2'-hydroxyl function, preferably until completion of the synthesis. In addition, it must be released under conditions mild enough to 14a R =
0
\/
14b R = - / T 6
/p\
0- 014
14c
RE 0
1%
/
R
15
\
R
II
R = R = -C-Ph
15b R = R =
y\
15dR=R= Figure 6. Protecting groups used for ribonucleotides.
\
/OCz",
C /' \cH,(c,H,)
4.
Protecting Groups
291
ensure no isomerization of the 3' 5'-phosphodiester linkage to the unnatural 2' 5' position. This means that any blocking group at this position must be stable to other manipulations required to extend the chain and hence its choice will determine the nature of all other protecting groups to be used in the synthesis. Three types of protecting group might be used: base-labile, acid-labile, or those labile to a specific conditions such as hydrogenolyzable benzyl ethers. Acetates or benzoatesa3 have been frequently employed as base-labile protecting groups in these positions, while the tetrahydropyranyl group (14a) is the most frequently encountered of the acid-labile blocking groups.44It has recently been shown that this group can be removed by 0.01 N aqueous ~ ~ conditions hydrochloric acid in 3-4 hours at room t e r n p e r a t ~ r e .These cause a negligible amount of phosphoryl migration. Various modifications of this latter group have also been reported, namely the 1-ethoxyethyl (14b)45 and the 4-methoxytetrahydropyran-4-yl ( 1 4 ~ derivatives. )~~ Frequently, by blocking the cis-diol system in the 2',3'-positions of the ribose molecule, the chain can be enlarged from the 5'- end. In addition to acetyl- and benzoyl-groups (lsa)," the base sensitive cyclic carbonates (15b)48 have been cited as potential blocking agents. Modifications to acidsensitive groups include substituted benzylidenes ( 1 5 ~ )or~ ~the alkoxy methylene derivatives (I5d) which can be converted into 2'- and 3'4-acyl derivatives by treat men t with acetic acid.5o -+
-+
E. For Phosphate The blocking of a terminal phosphate from further reaction either intramolecularly to form cyclic phosphates or intermolecularly to form pyrophosphates is absolutely essential in nucleic acid synthesis. Formation of a @-cyanoethyl ester (16s) remains probably the most frequently employed method.61It can be quantitatively introduced (0-cyanoethanol and DCC for 2 days) and is removed by momentary treatment with mild alkali. This extreme lability is also a drawback since, frequently, reprotection of the phosphate must occur after every step. The 0-elimination mechanism for its removal is typical of a method frequently encountered for deblocking the phasphate group. Thus esters of 2-acetyl-2-methylethanol (16b) are found to be more sensitive than 0-cyanoethyl derivative^,^^ whereas the 2-(a-pyridyl)-ethyl ester (16c) is most effectively removed by anhydrous sodium methoxide at 0°C after 48 hours.53 The @,@,P-trichloroethyl group (16d) has also been used in oligonucleotide synthesis and is removed reductively with Zn/Cu in acetic acid or dimethylf ~ r m a m i d e .However, ~~ these conditions are apparently not suitable for
292
The Total Synthesis of Nucleic Acids
N-benzoylcyt~sine.~~ A group of substituted phenyl hydracylamides (16e16f) has also been developed. These are appreciably more stable to mild alkali and consequently permit selective hydrolysis of the 3’-O-a~etyl.~~ A direct displacement can be achieved under acidic conditions by using benzhydryl (17a)” or r-butyl esters (17b)58;certain substituted phenols (17c) are susceptible to aqueous alkali treatment.5B 16 (removed by /?-elimination) 16a X = -OCH,CH,CN ” * 16b X = -OCH2CH-COCH3
0
I
CH3
OH.
1 6 ~ X = -OCHz-CH, 16d X = -OCH&CI, 16, 17, 18 16e X = -OCH,CHZCONHPh 16f X = -0CH~CtI~CONHCH2Ph 17 (removed by direct displacement) 18 (removed by specific reagent) 17a X = -OCH(Ph), 18a X = --S-C2H, 17b X = -O-C(CH3)3 lab X = -0CHzPh 17C X = -0Ph 1 8 ~X=-NH-
Figure 7. Blocking groups for nucleotide phosphate.
A number of blocking groups are removed only under specific conditions. Ethylthioesters (18a) are selectively removed by oxidation with aqueous iodine;65 this technique can also function as an activating procedure for oligonucleotide synthesis. Benzyl phosphates (18b) are split by hydrogenation,BObut this is often accompanied by reduction of the pyrimidine ring. However, benzyl phosphate triesters, formed by reaction of the internucleotidic phosphodiester bond, are debenzylated with sodium iodide in
5. Chemical Syntheses of Polynucleotides
293
acetonitrile at 80°C.s1Some recent investigations have shown that phosphoroanilidates (18c) can be specifically cleaved by isoamyl nitrite.62A 2',3'-cyclic phosphate has also been employed as a method of protecting both the 3'-phosphate and the 2'-hydroxyl while adding to the chain from the 5'hydroxyl end.6g Perhaps the most curious report concerns the use of a substituted uridine (18d) as a phosphate-protecting group for deoxyoligon u c l e o t i d e ~ Deblocking .~~ is accomplished in a stepwise fashion in which the key step is oxidation of the cis-diol by periodate. 5. CHEMICAL SYNTHESES OF POLYNUCLEOTIDES
In the last decade, most attempts directed at the synthesis of oligonucleotides have involved the concepts and reagents outlined in the preceding section. Several variations of two fundamental approaches, polymerization and stepwise condensation, have been used and are outlined next. A. Polymerization Method
Homopolymers
The availability of deoxyribopolynucleotides containing a single nucleotide unit aids in a variety of chemical, physicochemical, and enzymatic studies in the nucleic acid field. These compounds have been prepared by the polymerization of mononucleotides containing a free hydroxyl group. Initial studies, involving the reaction of thymidine 5'-phosphate with DCC,65 produced a large number of polymeric products. As a result, the major problem was the development of satisfactory chromatographic methods for separation and characterization of the desired linear polynucleotides, which were isolated mainly by chromatography on columns of cellulose anion exchanger (ECTEOLA) and diethylaminoethyl (DEAE) cellulose. Usually two homologous series of polynucleotides are obtained. The first are the linear polynucleotides represented by the general structure (19), while the second series of compounds contain the cyclic oligonucleotides represented by (20); they arise by an intramolecular phosphodiester bond formation between the 5'-phosphomonoester group and the 3'-hydroxyl group. The formation of the cyclopolynucleotide compounds can be reduced by adding some 25 % of 3'-O-acetylthymidine 5'-phosphate to the unprotected nucleotide at the start of polymerization.6s The 3'-OAc nucleotides form the terminating unit of the greater portion of the resulting polynucleotide chains thus blocking the cyclization reaction and the acetyl group can subsequently be removed by mild alkali treatment. By using this technique and working in concentrated solution (1 M), the major products obtained were the linear
294
The Total Synthesis of Nucleic Acids
0
I
0 - - - - - -/- - - - - -- - -O-P-OH
/
O==$-OH
-0
OH 11
n = 0-3
= 0-9
20
19
Figure 8. Major components (linear and cyclic oligonucleotides) of a polymerization reaction of pT. [Adapted from J. Atti. Cheni. SOC.,83, 675 (1961).]
polynucleotides, of which about 45 % of the total polymeric mixture consisted of linear polynucleotides longer than tetranucleotides. Purification of components up to dodecanucleotides on a fairly large preparative scale was attained on a DEAE-Cellulose column (carbonate form) by using volatile triethylammonium bicarbonate as eluent (see Fig. 9). The foregoing method for chemical polymerization by dicyclohexylcarbodiimide was also extended to N6-benzoyldeoxyadenosineS-phosphate,''*
Frrcilon numbr
Figure 9. Chromatography of thymidine polynucleotides (total polymeric mixture) on DEAE-Cellulose (bicarbonate) column. [Adapted from J . Ani. Cherri. Soc., 83, 675 (1961).]
5. Chemical Syntheses of Polynucleotides
295
N6-anisoyldeoxycytidine5'-ph0sphate,~'~and Na-acetyldeoxyguanosine 5'phosphate.67C Treatment of the reaction mixture with concentrated ammonia gave the expected homologous series of linear and cyclic polynucleotides. Other commonly encountered side products from this type of polymerization were compounds containing mono- and oligonucleotides linked to each other by pyrophosphate bonds (see 21). Their separation from the desired linear polynucleotides was extremely difficult; therefore before chromatography the entire polymerization reaction mixture was treated with excess acetic anhydride in pyridine to cleave the pyrophosphate bonds.68 Another series of minor contaminants was also present. The simplest member of the series contained a phosphomonoester group which could
22
Figure 10. Typical contaminants (pyrophosphates and pyridinium nucleotides) of a polymerization reaction.
be dephosphorylated by incubation with bacterial alkaline phosphatase. The product thus obtained had an ultraviolet spectrum similar to that of an equimolar mixture of thymidine and N-methylpyridinium cation. From these results, it was tentatively concluded that the cation had the structure (22), while the phosphomonoester group was located at the 3'-position. Therefore it was assumed that a homologous series containing pyridinium-substituted terminal nucleotides were also present. Various other reagents such as p-toluenesulfonyl chloride, 2,s-dimethylbenzenesulfonyl chloride, diphenylphosphorochloridate were also investigated but they did not compare favorably with DCC as a polymerization
1leu9 0
0
2 0 0 -10
9
21-
Figure 11. Two-dinlensional TLC resolution of chenlically polymerized pT on AvicelCellulose plate (20 x 20 cm, 0.1 mm thickness) (A) with Mesitylenesulfonyl chloride (MS) and (B) with dicyclohexylcarbodiimide (DCC). Solvent I, n-propanol: concentrated NH,:H,O (55:10:35) was used in the first dimension. Solvent 11, lsobutyric acid: I b f NH,OH:O.IM EDTA (100:60:1.6) was used in the second dimension. [Adapted from Bioc/ierfr. Eioplys. Res. C O w J i . , 41, 1248 (1970).] 296
(DCC)=B
L I
0 O G
297
298
The Total Synthesis of Nucieic Acids
reagent.68 I t has been reported that a polythymidylic acid of 30 units was obtained by using picryl chlorides9as the polymerizing reagent. Morerecently, /3-imidazolyl-4(5) propanoic acid has been found to act as a specific catalyst in the polymerization of the unprotected 5'-deoxyribopolynucleotides.7° Only the 3' 4 5' internucleotide diester bond is apparently formed and the overall yield of di- and oligonucleotides is reported to be about 50%. The products were contaminated with 30-40% of the 5'-dephosphorylated oligonucleotides. Treatment of thymidine 5'-S-ethylphosphothioate with iodine in pyridine in the absence of any external nucleophile produced extensive self-condensation by attack of the 3'-hydroxyl group on the iodineactivated phosph~rothioate.~~ Recently a thin-layer chromatographic technique has been employed for the separation and characterization of various products of polymerization of thymidine 5'-phosphate.'' This highly sensitive analytical technique clearly indicates not only the extensive polymerization occurring with MS and DCC treatment but also the various series of unwanted by-products (see Fig. I I). Block Polymers
For the synthesis of polynucleotides containing repeating dinucleotide units, an approach has been used which is similar to that already described. It involved the polymerization of suitably protected dinucleotides such as (d-pTpC"l'), (d-pTpG"C), (d-pCA1'pA"z)), and (d-A1'ZpGAc)with DCC in pyridine followed by the removal of protecting groups with alkali and separation of products by a combination of anion exchange and paper chromat~graphy.'~ In addition to the desired compounds, two main types of side products were identified : those bearing a 3'-phosphomonoester group at one end of the chain and a 5'-phosphomonoester at the other end (24), and those bearing 5'-phosphomonoester groups at both ends with an unusual C,.-C,. internucleotidic linkage within the chain (26). In all the contaminants, a common property which suggested the presence of two phosphomonoester groups per molecule was their striking increase in mobility on paper chromatography after treatment with phosphomonoesterase enzyme. Formation of these side products could be explained by postulating various intermolecular and intramolecular triesters or pyrophosphates (see 23,25, 27 in Fig. 12) as intermediates. These reactive triesters could then be broken down with pyridine from the reaction medium or hydroxide ion from the workup to give the observed products as well as the pyridinium-substituted nucleosides (28) observed earlier (see Fig. 10). There is no doubt that polymerization methods available at present are inefficient and produce a plethora of contaminating side products which require extensive chromatography. The ready formation of these side
5. Chemical Syntheses of Polynucleotidea 27
25
23
T
24
+22
299
P
+22
T
26
T
C
T
28
Figure 12. Contaminating side products from a polymerization of d-pTpC?lI, showing possible mechanisms.for formation. [Adapted from J . An?. Chm. Suc., 87, 2956 (1965).]
products illustrates very well the reactivity of the phosphate group and the multitude of sites available for reaction in these molecules. Nevertheless, this method does provide a rapid means of obtaining modest quantities of shortchain deoxyribopolynucleotides containing repeating sequences. For the extensive polymerization of the trinucleotides (d-pA”zpAuzpCA‘l), (d-pTpTpGaiC), (d-pTpA1’I.pGAC),(d-pA’”pTpC*”), (d-pA’i’pTpCA”), (dpA”ZpTpG“C), (d-pCA”pG~“pA”), (d-pC-illpG*cpT), (d-pCA1’pC-“’pT), (d-pG“CpG*CpAuZ), (d-pCA1lpCA1lpA1)Z), (d-pABzpAnZpGAc) and (d-pTpTpCAI1), MS was found to be a more effective reagent.’13Side products such as cyclic trinucleotides in minor amounts were always identified; this may be due to the use of excess condensing reagent. Also present were the usual contaminants containing phosphomonoester groups at both terminals and polynucleotides with a quaternary pyridine group at the 5‘-carbon of the terminal nucleotides. These methods were subsequently extended for the polymerization of tetranucleotides such as (~-pA”zpA”ZpA”zpG*c) and (d-pAuzpTpC“l‘ pG Ac). 74
300
The Total Synthesis of Nucleic Acids
B. Stepwise Condensation Methods for Polynucleotides without a Terminal Phosphate 1. The prevalent approach used in the stepwise synthesis of specific deoxyribopolynucleotides has involved the successive addition of mono-, di-, tri-, or tetranucleotide units to the 3’-hydroxyl end of a 5’-0-trityl protected mono- or oligonucleotide, as illustrated in Fig. 13. As the chain enlarges, an increasing excess of the incoming 3‘-O-acetyl protected unit (29) or (30) must be used to maintain satisfactory yields with respect to the “growing” chain. The condensing agents in current use are MS, TPS,or DCC. In most of the synthetic work MS or TPS reagents offer the particular advantage that trialkylammonium salts can be used to aid solubilization of the nucleotidic components in the reaction medium.75 Moreover, the presence of a trace of amine is a strong inhibitor of the DCC reaction. After each condensation step, the reaction mixture was submitted to a mild alkali treatment to remove the 3I-O-acetyl group and the desired product was obtained by anionexchange chromatography on a DEAE-Cellulose column using volatile triethylammonium bicarbonate pH 7.5 buffer. By using this approach, the synthesis ofa chain upto hexadecanucleotide (TpTpApC),has been achieved.76
-0
Q I
CH ,C=O 30
Figure 13. A representative stepwise condensation to prepare an oligonucleotide without a terminal phosphomonoester group. 301
302
W
W 0
I
I
0
II
0
I
O
q
"=I
O=CCH 2CH 2CCGH 9
OQT
0
I
0
O=P
I I O
T
31
q T OH
Figure 14. The triester approach to oligonucleotide synthesis. [Adapted from J . Am. Chem. Soe., 91, 3360 (1%9).]
0 OQ O=CC T 0 I H ,CH ,CC,,H II
I
H O T *
v,
MTroQ 0 I O=POCH~CH~CN I I O T
304
The Total Synthesis of Nucleic Acids
Side products were formed in small amounts in every condensation reaction, Two homologous series of compounds were positively identified.77 The first group of compounds were Tr-Tp, Tr-TpTp, and Tr-TpTpTp in the thymidine series; these correspond to products already discussed in the polymerization reactions, that is, a series in which the 3’-hydroxyl has been phosphorylated (see 24), and another series containing pyridine substituted at the 5‘-carbon of the sugar moiety (see 22 and 28). In addition, stripping the anion-exchange column with 1 M salt at the end of gradient elution invariably yielded 5-20 % of the total nucleotidic material. The formation of these uncharacterized “side products” with apparently a higher number of charges than those in the desired product is more serious; they may arise by phosphorylation of the heterocyclic ring by the activated n u c l e ~ t i d e s . ~ ~ This approach has two main disadvantages: (a) To maintain high yields in the condensation step, increasingly large excesses of the 5’-O-phosphomonoester component must be employed as the chain length of the oligonucleotide is increased. This requirement makes the approach very uneconomical from a practical point of view. ( 6 ) At each stage in the synthesis the products must be separated as salts by chromatography on DEAECellulose columns with aqueous buffer solutions. Although efficient, such chromatography is time consuming (a typical separation requires several days) and yields the high molecular weight material-desired products-at the end of the elution pattern. However, this approach has been solely employed in the synthesis of all the deoxypolynucleotides required for building the gene corresponding to alanine rRNA.’O 2. A synthetic approach involving 0-cyanoethyl phosphotriesters for the synthesis of oligodeoxyribonucleotides has recently been reported.*O As uncharged molecules, the phosphotriesters produced as intermediates would be expected to be soluble in organic solvents and amenable to the conventional techniques of separation and characterization of organic molecules. In addition, masking the active oxygen in the phosphodiester groups should prevent formation of the pyrophosphate bonds which are apparently responsible for chain fission and unwanted side products. The synthetic scheme, as outlined in Fig. 14, involves phosphorylation of the 3’-hydroxyl of a nucleoside or oligonucleotide derivative using pyridinium mono-(p-cyanoethyl) phosphate and MS or TPS. This phosphorylation is followed by condensation of the phosphodiester with an appropriate nucleoside, again employing an arenesulfonyl chloride as condensing agent. Workup of this mixture and chromatography on a silica-gel column with ethyl acetate and tetrahydrofuran afforded the desired product (e.g. 31), after removal of protecting groups. By using this method the synthesis of a hexanucleotide, (pT),, has been achieved.34 A trichloroethyl ester of phosphate has also been reported in a
5. Chemical Syntheses of Polynucleotides
305
phosphotriester approach;81 also, benzyP and phenylBaesters have been mentioned in triester-type syntheses.
3. An attempt has been made to prepare deoxyribooligonucleotides by activating the phosphate group to nucleophilic attack by a free hydroxyl which is ionized by strong base.’* Only a modest yield of a tetranucleotide DMTr-TpTpTpT-MMTr was obtained, but it is interesting to note that no protecting groups were required for the amino functions and pyrophosphatecontaining products were not detected. However, two major problems must be solved: appreciable cleavage of the glycosidic bond in the strong anhydrous base and the limited solubility of the oligonucleotide salts in dimethylformamide. C. Stepwise Synthesis of Deoxyribopolynucleotides Bearing a 5’-Phosphomonoester End Group 1. For the synthesis of longer deoxypolynucleotides, it is necessary to have oligonucleotide units containing a terminal phosphomonoester end group, usually at the 5‘-position. Chemically, such compounds have been synthesized by the stepwise condensation of a P-cyanoethyl protected mononucleotide with a 3’-O-acetyl mononucleoside 5’-pho~phate.~~ The product was isolated by DEAE-Cellulose column chromatography, reprotected with a P-cyanoethyl group, and condensed with another appropriate mononucleotide, as outlined in Fig. 15. Insertion of phosphate at this position can also be accomplished enzymatically under certain conditions (see later section on enzymes). 2. By approximately doubling the oligonucleotide chain during each condensation step (see Fig. 16) the sequential method has been modified and extended for the synthesis of various dodecan~cleotides.~~ In this approach, products and reactants differ substantially in molecular weight and therefore can be separated rapidly (products are usually off the column within 24 hours) and quantitatively by gel filtration through Sephadex gels with appropriate exclusion limits. An attractive feature of this separation technique is that the product peak emerges from the column before the starting material (see Fig. 17). However, the presence of the extremely labile 0-cyanoethyl ester necessitated reblocking of the 5‘-phosphate at each stage. In principle this difficulty could be overcome by using the recently developed P-0-/3-trichloroethyl ester, S-ethyl phosphothioate, or anilidate-protecting groups which are more stable but can be cleaved selectively.
3. Recently a new approach, which utilizes substituted benzene derivatives (a) to block the phosphate group and (6) to increase dramatically the binding property of the nucleotidic fragment to benzoylated DEAE-Cellulose, has
306
The Total Synthesis of Nucleic Acids
been developed for the synthesis of deoxytibooligonucleotides of defined sequence.66*59Hence after a condensation step the reaction mixture is passed through a column of benzoylated DEAE-Cellulose or benzoylated DEAESephade~.~5 This technique offers the distinct advantage of effecting the complete removal of any products lacking aromatic residues. Thus, if nonaromatic protecting groups are used for other functionalities, a facile separation of unwanted contaminants such as pyrophosphates, cyclic phosphates, and the starting material containing the free phosphate is assured. The most useful blocking groups are phenyl-substituted hydracrylamides, which, 0
II
OH
+
0 0 - - J O C H q
II
/O 0-P=O / ocm’
I)
DCC
v 2) OH
0
II
-A
0 R
NCCH2CH,0-POCH,
(a
/O
0-r-0 /
OH
O C W OH 31a
Figure 15. The stepwise approach to the synthesis of deoxyribooligonucleotidecontaining a
5‘-phosphate.[Adapted from J. Anr. Chem. Soc., 89, 2158 (1967).]
0
31a
+
0
1) rncsllvlenesulronvl chlorldc or
lrlisopropylknmne sulfonvl chloride 2)
-0
b
6H
/”
0-P=O /
OH Figure 15.
32
(Corttinued)
032b 32a
+ 32b
1) mesitvlenesulfonyl
chloride
2)
on-
t
hexanucleoside 5’-phosphate
3 ) gel fillralion on Scphadex 0 - 7 5 (Superhe)
/3-cyanoethyl ester of hexanucleotide
+
3’-O-acetyl hexanucleoside 5’-phosphate
1) meslcylcncsul~onvl chloride
:
+ dodecanucleoside 5’-phosphate
::&alion on Sephadcx G-75 (Superfine)
Figure 16. Block synthesis of deoxyoligonucleotides possessing a S’-phosphate group. 307
308
The Total Synthesis of Nucleic Acids
36.0
-
34.0
-
32.0
-
30.0
-
- 28.0 2
-
8
24.0
-
I
22.0
-
I
-
-
26.0
hl
" " l ' l ' " I '
.. E' 20.0 -
I- :I 0
5 e U
14.0
-
2P 12.0 -
8.0 -
10.3
6.0
-
4.0
-
2.0
-
0
40
60
80
l l I l I l l l l l l l I 1 1
I
1
100 120 140 160 180 200 220 240 260
Fraction number
Chromatography of the reaction mixture (one-half portion) on a Sephadex G-75 (superfine), K25-100 column in the preparation of d-pABzpCAnpTpABzpCAnpAn". Fractions of 2 ml were collected every I5 min. Fractions 108-128 contained the desired product. [Adapted from Biocltem., 8, 3443 (1969).]
Figure 17.
unlike the P-cyanoethyl group, are stable to conditions necessary for removing the 3'-O-acetyl group. Consequently, the phosphate does not have to be reprotected before each new condensation step, a substantial saving in time. Preparative scale TLC techniques on Avicel cellulose ( I .O mm thick) can also save much time in the isolation of synthetically useful amounts of pure
oligonucleotide^.^^
5. Chemical Syntheses of Polynucleotides
309
D. Polymer-Support Synthesis
Some analogies can be drawn between the problems involved in nucleic acid synthesis and those of peptide synthesis. One of these is the necessity of rapid purification of product after each condensation step, a problem which has been elegantly circumvented in peptide work by Merrifield in the so-called solid-phase or polymer-support synthesis. Essentially, the method consists of attaching the growing chain to an insoluble and inert carrier by a covalent bond. After each condensation, the product is freed from incoming soluble component and reagents by simple filtration, thus eliminating any timeconsuming chromatographic separations. During the last few years, several reports have appeared describing initial attempts at polymer-supported deoxyribooligonucleotide synthesis. Two of the special problems associated with these investigations are the polar nature of the nucleotides (necessitating polar solvents) and the extra functional groups that have to be protected. Polystyrene supports having some reactive sites for attachment of the nucleotide (see below) have been used. Two general types have been reported : (a) soluble in the reaction medium, which is often anhydrous pyridine or dimethylformamide, and precipitated from water which is then used to wash out the reagents;8Eand (b) insoluble in the reaction medium.87These polymers are generally crosslinked with some divinyl benzene and, depending on the amount of crosslinking, they often swell, thus making elution of reagents more difficult. Every functional group in the nucleotide has been used as a point of attachment to the polymer (see Fig. 18). Obviously the polymer will have some type of active site involving benzene rings and covalent linkages have been formed via: 1. A trityl (or derivative) ether bond to the 5'-hydroxyl of the nucleotide, 33.88
2. A benzoic ester bond to the 5'- or 3'-hydroxyl, 34 and 36 respectively.88 3. A benzamide bond to the primary amine of the heterocyclic base, 37.80 4. A phosphoramidate linkage to the 5'-phosphate, 3Sg1
Both the diester and triester approach have been used to form the phosphate linkage between two nucleotides and all four different nucleosides have been employed in the investigations cited to date. With these methods deoxyribooligonucleotides of up to six units88b have been prepared but, depending on the conditions employed, various problems have been encountered. Swelling of some supports often makes it difficult to remove all traces of starting material and reagents after each condensation step. The more insoluble polymers of course lead to heterogeneous reaction mixtures which
310
The Total Synthesis of Nucleic Acids
can give rise to lower yields in the condensation step or alternately require longer than usual reaction times or more vigorous conditions. To remove the oligonucleotide chain from the support polymer frequently requires very vigorous conditions, and this difficulty increases with increasing chain lengths. A recent report describes the synthesis of hexa dT on an insolubletype support in one week.02 Many of the problems mentioned were apparently overcome; however, a fiftyfold excess of incoming mononucleotide was
0
I
Cl
33
1) CH CONHCH,CH,O2) H&
CH,-0-CH,-CH,-NH,
OH
-
34
pTIDCC
0 @-NH-P-o
II
0 OH 35 Figure 18. Preparalion of various polymer supports and their attachment to the nucleotide.
5. Chemical Syntheses of Polynucleotides
311
Tt
37
Figure 18.
(Conrimred)
required at each condensation step and the experimental details were very sparse. To be distinctly advantageous, the polymer support method requires a fast, almost quantitative condensation step and ready solubility of reagents. Both of these difficulties remain in nucleic acid syntheses. E. Synthesis of Oligoribonucleotides
The synthesis of ribooligonucleotides is greatly complicated by the presence of the 2’-hydroxyl function on the ribose moiety. Since the natural phosphate bridge occurs between the 3‘- and 5’-positions of two adjacent nucleosides, one must be able to differentiate between the two hydroxyl functions at the 2’- and 3’-positions, both of which are secondary. Additionally, the 2’hydroxyl should have a protecting group that is not removed under conditions that deblock either the 3‘- or 5’-positions prior to condensation; however, it must be readily deblocked at the conclusion of a synthesis and under conditions that will not isomerize the 3‘ -+ 5’ linkage to 2‘ 4 5‘. It has been shown that either acidic or basic conditions can accomplish this migration to the “unnatural” linkage. Although some small homopolymers have been synthesized from suitably protected monoribonucleotidese3and some attempts have been made to achieve syntheses from cyclonucleoside precursors,B4the
312
The Total Synthesis of Nucleic Acids
main effort in the ribo-field has been directed toward a stepwise synthesis of chains containing defined sequences.95 Several general approaches can be conceived and two of the most thoroughly investigated are illustrated in Fig. 19. Initially, let us assume that any primary amino functions in the purine or pyrimidine rings can be satisfactorily protected and that there is no
HO OR’ 41
t
38
Ri)
RO OR
bR
39
Figure 19. Two generalized
40
42
approaches to the synthesis of oligoribonucleotides.
difference between effecting the internucleotide bond via a condensation of 3’-phosphate to 5‘-hydroxyl or 3’-hydroxyl with 5‘-phosphate. There are no known chemical methods for quantitatively distinguishing the 2’- and 3’-positions; however, with the aid of purification techniques compounds such as 38 and 41 can be obtained in pure form. Early efforts were focused on preparing 38 from selective, but not specific, basic hydrolysis of the 3’-5’-cyclic phosphate. Differentiation was achieved by using trityl at the 5’-position and tetrahydropyranyl for the 2 ’ - h y d r o ~ y l . This ~ ~ was later modified to the 5’-OTr and 2‘-OA~.~31, The use of pancreatic ribonuclease (RNase) to specifically and quantitatively cleave 2’,3’-cyclic phosphates to the 3’-phosphate, 2’-hydroxyl substitution pattern is exemplified by the preparation of 5‘-OAc, 2’-OTHP n u ~ l e o t i d e s . ~Other ~ ” vinyl ethers were introduced at the 2’-position to make deblocking easier.45 In order to be able to extend the chain, the 2’,3’-cis diol system of the entering nucleoside component, 39 must be protected with a group that is completely stable under conditions used for deblocking the 5’-hydroxyl. Thus acid-stable (acetyl, benzoyl) groups have been employed when trityl or derivatives are used to block the 5’-hydroxyl,4’ whereas base-stable alkylidene derivatives are employed when 5‘4-acetyl is present.44
6. Enzymes as Reagents
313
The second approach, 41 plus 42, requires differentiation of the 2’- and 3’-hydroxyls in the nucleoside component 41 and has been partially achieved by purely chemical method^.^^'^' Although isomerically pure samples could be isolated by crystallization techniques, the overall yield suffers from a somewhat lengthy sequence. However, several interesting points have been reported. The THP group protecting the 2‘-hydroxyl can be removed under acidic conditions so mild that negligible isomerization to 2‘-5’ occurs on deblocking. Furthermore, the bulky, pivaloyl group affords essentially specific protection for the primary 5‘-hydroxyl and is base-labile (acid-stable) in contrast to the commonly used trityl group. Both of these procedures suffer from the same drawbacks. The final product 40 has no terminal phosphate, which would be necessary for chain elongation. If this discrepancy could be overcome, say by chemical or enzymatic phosphorylation, the chain could only be extended from the 5’- end since again the 2’- and 3‘-hydroxyls of the 3‘-terminus could not be specifically differentiated. An attempt to overcome these difficulties was reported in which the first route, 38 39 + 40, was used; however, the 2‘- and 3’-hydroxyl groups of the nucleoside were “protected” by using the readily available 2’,3’-cyclic p h o ~ p h a t eCondensation .~~ and workup produced a dinucleotide diphosphate (mixture of 2’- and 3’-phosphates at the 3’-terminus), which could be quantitatively cyclized with DCC and then specifically reopened to the 3‘-phosphate using RNase. The chain was immune to enzymatic degradation since all the internal 2’-hydroxyl groups were protected. Two major problems presented themselves. Although unwanted side products due to polymerization of the cyclic phosphate could be removed by chromatography, they presumably reduced the yield of the desired product. The enzymatic cleavage was practically useful only when no protecting groups were present on the heterocyclic base at the 3‘4erminus. Therefore either that base must have no free amino group such as uridine, or it must be specifically deblocked-without affecting any other protecting groupsbefore the enzyme can function.
+
6. ENZYMES AS REAGENTS
The search for a more complete understanding, at the molecular level, of the mechanisms involved in the various reactions of DNA, for example, replication and transcription, has been greatly influenced by the various oligonucleotides fabricated by organic chemists. Paralleling these advances were the isolation and investigation of several enzymes. The interplay of these two factors has advanced the understanding of this vital biological
314
The Total Synlhesis of Nucleic Acids
process at an incredible rate during the last decade. In addition, the ready availability of several of these enzymes in pure form has allowed them to be used as routine-and highly specific-reagents for the synthesis and characterization of polynucleotides. DNA-Polymerase
A.
An enzyme has been isolated from E. colie6 which catalyzes the synthesis of high-molecular-weight DNA in the presence of all four of the deoxynucleoside 5'-triphosphates, d-ATP, d-CTP, d-GTP, and TTP, plus a DNA template. No synthesis of DNA takes place in the absence of this template, which may consist of short-chain synthetic oligonucleotides or longer chains isolated from natural sources. For example, it has been demonstrated that short-chain synthetic deoxyribopolynucleotides containing alternating deoxyadenylate and thymidylate units serves as a template, in the presence of thymidine 5'-triphosphate and deoxyadenosine 5'-triphosphates, to induce the synthesis of a high-molecular-weight DNA-like polymer containing again deoxyadenylate and thymidylate units in alternating sequences (see Table 1 , Reaction I).Table 1 lists the types of reactions so far elicited from DNA-polymerase. These reactions are also very fast; polymers much larger than the original template can be produced in several hours. Table 1. DNA-Polymerase Catalyzed Reactions
Reaction
Reference
+
1. d-(AT), + d-ATP TTP 2. TI, + d-A, d-ATP TTP 3. TI, + d-A, + d-ATP
+
4. d-(TG),
+ d-(AC), +
5 . d-(TTC),
+ (I-(AAG),
6. d-(TATC),
+
+
POIYd-AT POIYd-A :T --c POIYd-A + +
97 98 98
POIYd-TG :CAR
98
-+
POIYd-TTC:GAA"
99
---c
POIYd-TATC:GATA"
+
100
a All of the DNA-like polymers are written so that the colon separates the two complementary strands. The complementary sequences in the individual strands are written so that antiparallel base-pairing is evident.
6. Enzymes as Reagents
315
Table 2. Nearest Neighbor Frequency Analysis of Poly d-TTC:GAA Templates: d(TTC), d(AAG)s
+
Radioactivity in Deoxynucleoside 3'-Phosphates dAP
dGP
a-P3'-Labeled Triphosphate
Count/ Percentage Min
dATP dGTP dCTP
12.836 13,684
dTTP
0
0
50 100 0 0
dCP
Count/ Percentage Min
Count/ Min
12,851
50
0
0
0 0
0 12,860
0 0
0
0
dTP
Percent- Count/ Percentage Min age 0
0
0 50.6
0
0
9,623 12,565
0
0
100 49.4
The most direct method for showing the chemical structure of the polymer is by the so-called nearest neighbor analysis technique,'O' in which one a-P32 labeled deoxyribonucleoside triphosphate is incorporated into the chain at a time. Degradation of the synthetic polymer to deoxyribonucleoside 3'-phosphates using enzymatic methods gives results which show complete accord with theoretical expectations from the sequence of the original template. Typically, the results from such a polymerization using templates with complementary repeating trinucleotide sequences is shown in Table 2.89 Within experimental error, excellent agreement is observed concerning the incorporation of the various 5'-triphosphate monomers. An additional property of these synthetic DNA-like polymers is their ability to reseed the synthesis by the DNA polymerase of more of the same The importance of this finding can hardly be overstressed. In essence, it means that once the specific sequences have been put together by well-defined and unambiguous chemical synthesis, DNA-polymerase will ensure their permanent availability. Thus the well-known, dramatic feature of DNA-structure, its ability to guide its own replication, can be exploited at the molecular level. The total number of DNA-like polymers prepared so far is listed in Table 3. Table 3. New DNA-like Polymers with Repeating Sequences
Repeating
Dinucleotide
Sequences
POIYd-TC: AG P01y d-TG A C
Repeating Trinucleotide Sequences
Tetranucleotide
Repeating
POIY(I-TTC:GAA POIYd-TTG :CAA POIYd-TAC:GTA POIY(I-ATC:GAT
POly d-TTAC: GTAA POIYd-TATC:GATA
Sequences
316
The Total Synthesis of Nucleic Acids
The complete characteristics of the DNA-polymerase-catalyzed reactions can be summarized as follows : (a) chemically synthesized segments corresponding to both strands are required for reaction to proceed: (b) minimal size of the two complementary segments used as primers varies between 8 and 12 nucleotide units; (c) synthesis is extensive; ( d ) products are high molecular weight and are double stranded with sharp melting transitions ; (e) nearest-neighbor analysis invariably indicates that the individual strands contain the appropriate repeating sequences; (f) high-molecular-weight products can be reutilized as primers for more synthesis. An electron micrograph of poly d-TG:AC showed the average size to be 0.5 p. This is indicative of a molecular weight in the range of one million.Q8h B.
Nucleotidyl Transferase Enzyme
A second DNA-polymerizing enzyme, the “end-addition enzyme,” has been detected in a purified extract of calf thymus. Chromatography on a hydroxylapatite column permits its separation from DNA-dependent DNA-polymerase.lo2This particular enzyme extends oligodeoxyribonucleotides via the repeated addition of deoxyribonucleotide 5’-phosphate units to long homologous chains: pTpTpT
+ nd-ATP
end-addllion enzyme
pTpTpTpApApApA
* * *
The rate of this polymerization decreases in the sequence d-ATP > d-ITP > d-CTP >> d-TTP and d-GTP, while d-CTP is incorporated only in the presence of Co2+ions.’03 The primer oligonucleotides must have a minimum length of three nucleotides at the same concentration of enzyme, while the length of the newly formed chain depends on the substratelprimer ratio. Thus DNA-like polymers having specific sequences at one end can be obtained for biological testing.
C. RNA-Polymerase (DNA-Dependent) A DNA-dependent RNA-polymerase ,is present in bacteria and other tissues.lo4I t is similar in action to DNA-polymerase and requires a doublestranded segment of DNA as a primer plus all four ribonucleoside 5‘triphosphates to be effective. The polyribonucleotides thus synthesized are single-stranded and of high molecular weight and their composition is determined by the composition of the DNA template. If synthetic polydeoxyribonucleotides with a known base sequence are used as templates, then a polyribonucleotide having a complementary defined base sequence is obtained. Typical results obtained with various synthetic polydeoxyribonucleotide primers for DNA-dependent RNA-polymerase are given in Table 4.
6. Enzymes as Reagents
317
Such results emphasize the fact that only by providing in the reaction mixture a set of ribonucleotide 5'-phosphates which are appropriate for copying the template can the required single-stranded polyribonucleotide be obtained. In every case the single-stranded ribopolynucleotides thus prepared have been shown to contain the expected repeating sequences by the technique of nearest-neighbor analysis. A template of only 9 nucleotides (3 triplets) has Table 4. DNA-Dependent RNA-Polymerase Reactions
+ ATP
CTP
POIY ~-TG:Ac--(
\
+ Poly CA
RNA Polymerase
GTP
+ UTP
+
UTP + ATP + GTP poiy CTAC: G T A ~ RNA Polvmerase \ UTP + ATP + CTP
+
Poly d-TATC:GATA
Poly G U
poly GUA
Poly UAC
+
UTP ATP GTP RNA Polymerase UTP + AT + PCTP
-f
poly GAUA Poly UAUC
been shown to be sufficient for synthesizing a complementary polyribonucleotide with more than 150 nucleotide units.los D. Polynucleotide Phosphorylase
Polynucleotide phosphorylase polymerizes nucleoside diphosphates to polynucleotides in which the bases are distributed at random: n X D P e (XMP),
+ nPi
This enzyme requires no templateIo6and is specific for ribonucleoside 5'diphosphate, even chemically modified 5'-diphosphates can be incorp~rated.'~' If several nucleoside 5'-diphosphates are present, they are incorporated into the polymer at random with the exception of GDP, which is incorporated preferentially.108 The nonspecificity of polynucleotide phosphorylase frequently can be circumvented by the action of other enzymes for the preparation of some definite characteristics. For example, random incorporation of a specific
318
The Total Synthesis of Nucleic Acids
nucleotide into a polymer and subsequent cleavage at this nucleotide can give a polymer with a definite triplet at the 3'-end:loe UDP
+ GDP (30: 1)
1 pUpUpGpU . pUpUpUpGpUpU .1 - pUpUpUpGp + pupu pupupcp + u 1 pUpUpUpG polynuclcolidc phosphorylasc
* *
* * *
*
a
T,rlbonuclcasc
*
* *
phosphstssc
* *
Moreover, this enzyme can be used for the synthesis of short chains. With the primer-dependent enzyme from M. lysodeicticus, a dinucleotide phosphate is extended by only a few units if the reaction takes place in 0.4 M NaCI.llo If the reaction is allowed to proceed until the few long chains formed initially have been converted by phosphorolysis into a large number of short chains, then the following sequence can be obtained: ApU
+ ADP ( I :4)
1 A (high molecular weight) + ApU (unused primer) polynuclcolidc phosphorylarc(2 hr)
ApUpApA
* * *
~polynuclmtidcphoiphorylare (24 hr)
+
+
ApU (small amount) ApUpA (main product) ApUpApA ApUpApApApA a small amount of polynucleotide of high molecular weight
+
+
The trinucleoside diphosphates XpYpZ can be separated chromatographically and thus is a purely enzymatic method suitable in principle for the synthesis of all 64 ribonucleotide triplets, which have been synthesized ~hemically.~7 E.
Ribonuclease Enzymes
Pancreatic ribonuclease has been mentioned as a reagent for opening 2',3'cyclic phosphates quantitatively to 3'-phosphate, 2'-hydroxyl in pyrimidinecontaining ribonucleotides, U or C. This reaction is only one step in a sequence of reactions defining the ribonuclease's real function as an endonuclease, that is, to split any phosphodiester bond specifically to yield a pyrimidine 3'-phosphate. Thus, the enzyme has also received extensive use
Enzymes as Reagents
6.
319
as a degradative tool in the sequencing of tRNA;lll for example, UpGpApCpUpUpApCp
-
* *
ipancrcalis R N ~ S ~
Up
+ GpApCp + Up + Up + ApCp
*
Additionally, TI ribonucleaselle is specific for guanosine and several other RNase have been isolated with other characteristics. Also, these enzymes appear to require a free 2‘-hydroxyl as well as a free amino function on the heterocyclic ring before they can be effective. Not all of them are commercially available and hence their use is limited. F. Polynucleotide Kinase
Although introduction of phosphate can be accomplished chemically by using many types of phosphorylating agents, the recently purified1l3 polynucleotide kinase can accomplish introduction specifically at the 5’-hydroxyl using only ATP. However, this enzyme requires the presence of phosphate in the 3’-position also, that is, a 3’-nucleotide or oligonucleotide chain, and is not commercially available. Consequently it has been employed only in a very limited sense often for incorporation of P3afor a crucial labeling experiment. G. Alkaline Phosphatase (E. Cofi)
This enzyme is used extensively for the nonspecific removal of any phosphomonoester group at alkaline pH.l14
H. Phosphodiesterases (Spleen and Venom) A phosphodiesterase is an enzyme which hydrolyzes the phosphodiester bonds linking the nucleotides in RNA or DNA. These two types of enzyme are commercially available and used routinely to characterize deoxy (or ribo) oligonucleotides. Venom diesterase hydrolyzes RNA to 5’-monon~cleotides~~~ and is also active in hydrolyzing the oligonucleotides produced by the action of deoxyribonucleases(1)on DNA to eventually yield deoxyribonucleoside 5’-phosphates. Spleen diesterase hydrolyzes RNA to 3‘-mononucleotides and also acts on the mixture of oligonucleotides produced from DNA by spleen deoxyribonuclease(I1) to yield deoxyribonucleoside 3’-phosphates.l16
320
The Total Synthesis of Nucleic Acids
I.
Polynucleotide Ligase
Recently, enzymes which catalyze the covalent joining of breaks in a single strand of bihelical DNA have been identified and purified from E. coli, phage-T, and phage-T, infected E. c ~ l i Given . ~ ~ a~ DNA substrate containing single-stranded breaks, as shown below, the ligase enzyme accomplishes the repair by the esterification of an internally located 3'-hydroxyl group with the adjacent 5'-phosphomonoester. In the reaction ATP is cleaved to AMP and PPi. 3'
...
I 1
A G T C
I
I
T C A G
. . .5' +ATP ___f
I I I I
5'
...
+
A G T C AMP T C A G P P P P P***3'
ssss
+ PPi
Studies on E. coli and T,-induced ligases have revealed that they have many properties in common. Both enzymes are specific for the same DNA substrates and produce the same DNA end product. Furthermore, the reactions catalyzed by both enzymes are rnediated by an enzyme-adenylate (enzyme AMP) complex. However, a distinguishing feature of the two enzymes is that the adenylate moiety for each of the two complexes is derived from different cofactors. The cofactor for E. coli enzyme is DPN, whereas that for the T,-induced enzyme has been shown to be ATP. Much has been deduced concerning the overall mechanism of the reaction. In both cases, the first step in the overall reaction consists of the transfer of an adenylate group from the cofactor to the enzyme to form a covalently linked enzyme-AMP intermediate. Once enzyme-AMP is formed by a reaction of the enzyme with DPN, there is a further transfer of the AMP to the 5'-phosphoryl terminus of a DNA chain to generate a new pyrophosphate bond linking the AMP and DNA. In the final step, the DNA phosphate of the pyrophosphate is presumably attacked by the 3'-hydroxyl group of the neighboring DNA chain, displacing the activating AMP and
7. Gene Synthesis
321
forming the phosphodiester bond.118However, specific characteristics made these enzymes of singular importance to the problem of the total synthesis of biologically specific high-molecular-weight DNA. Initial studies showed that a short deoxyribopolynucleotide carrying a 5’-phosphate group at one end and a 3‘-hydroxyl group at the opposite end could be joined end-to-end in the presence of long complementary deoxyribopolynucleotides.llBThese experiments also showed that chains as short as hexa- or heptanucleotides could be joined. Further, if both the complementary polynucleotides are short, such as the hexadecanucleotide d(TTAC),, the octanucleotide dPs2(TAAG), can be added.
7. GENE SYNTHESIS The fruition of many of the observations outlined in previous sections of this review were realized by the dramatic announcement, in 1970, of the total synthesis of the gene for an alanine transfer RNA from yeast1e0 (see also reference 79). In defining the DNA sequence complementary to the t RNA sequence, the principle assumption was made that all the minor bases were produced from the four parent bases by modifications which occurred after transcription of the DNA gene with the four standard bases. The outstanding concept in the synthesis was to use polynucleotide ligase for joining relatively short, chemically synthesized, polynucleotide chains while they were held together in properly aligned bihelical complexes. The total plan for the synthesis of this gene is outlined in Fig. 20. The gene was divided into three parts, shown as A, B, and C (or C’),and each part was to consist of several chemically synthesized segments. The chemically synthesized segments were phosphorylated using T4-polynucleotide kinase, after heating to overcome any structural inhibitions of the phosphorylation reaction. The chains were elongated by stepwise addition of the appropriate segments using polynucleotide ligase. These segments were anchored in place by base-pairing with the overlapping end of the growing gene. The product from each addition could be separated from the starting materials on Agarose or Sephadex columns. The extent of the ligase-catalyzed joining reactions ranged from 40 to 70%. An analysis of the desired product hinged on the following steps: (a) resistance to phosphatase; (b) degradation to 3’-nucleotides using micrococcal and spleen phosphodiesterase; ( c ) hydrolysis to 5’-nucleotides using pancreatic deoxyribonuclease and venom phosphodiesterase. In a few cases repair with DNA polymerase was also used as further characterization.121
U U C
1
1
1
1
1
1
C U
1
1
C G
8
6
1
1
1
4 C A
5
C-C-A-C-C-A
1
U C
7
1 ' ) 1 T-G-A-G-C-A-G--G-T-G-G-f
C-C-G-G-A-C-T-C-G-T I 3 \ " ' 2 ) -
1
1
C G G A
9
1
C
3
C
1
2
A
1
1
Me*
C U
1
1
1
U
4-A-A-T-C
C
I
1
1
I
1
Me G G G A G A G U
y
1
1
1
1
u'75(;-)
1
1
1
1
1
C
1
1
U C
C G G
Hr
A-G-A-G-T-C-T-C-C-G-G-T-T-C-G-A-T-T
1
(6) G-T-A-C-C-C-T-C-T-C-A-G-A-G-G-C-C-A-A--G-
G C
-G-C-T-C-C-C-T-T-A-G-C-A-T--G--G-G
C U C
T
w
1
C G A
U U
50 49 48 47 46 45 44 43 42 41 40 39 38 37 36 35 34 33 32 31 30 29 28 21 26 25 24 23 22 21 20 19 I8 17
-G
'
A
1 4 , -C-T-A-A--G--G-C-C
G
20 19 18 17 16 15 14 13 12 1 1 10
(3') deoxy
(5') deoxy
(3')ribo
(3') deoxy
(5') dmxy
(3')nbo
B
A
w w N
C G
Me
1
1
1
1
1
1
1
1
1
1
W
1
1
1 - ( 1 3 ) 1
1
1
1
1
1
1
1
1
1
I
(
1
1
)
1
1
1
C G
A
1
O
C G )
Me, C U
T-C-G-C4-C-G-A-G-G-
V
A G
T-C44-T-A-G-C-G-C-
1
G-C-A-T-C-A--G-C-C-A
T--G--G-C--G-C--G-T-A-G
1
4
Hz
C G G U C
C-
1
1
1
1
C G
Me
1
1
1
1
1
1
1
1
h
(
1
3
T44-C--G-C--G-T-A4
1
1
)
1
U
1
H*
1
1
1
C G G U
1
1
A G
-
(
I
l
)
A
T-C-G-4-T-A-4-C-G-C-
L J
1
G
1
Me*
C G
C U
1
1
1
G-C4-C4-A4-G-
~ ( l W - 1
C G
C
C-
7
(3) deoxy
(5’) deoxy
(3’)ribo
(3‘) deoxy
(5‘) deoxy
(3‘)ribo
Figure 20. Total plan for the synthesis of a yeast alanine rRNA gene. The chemically synthesized segments are in brackets, the serial number of the segment being shown within the brackets. A total of 17 segments (including 10’ and 12’) varying in chain length from penta to icosanucleotides were synthesized. [Adapted from Nufure, 227, 27 (1970).]
I - < l 5 ) 2
G--G4-C-G-T4
1
A
H Z
~ ( 1 2 ’ ) . - , G-C-A-T-C-A-G-C-C-A-T-C
U G U G G C G C G U
C-C-C--G-C-A-C-A-C-CC--G-C
W -1
G G G
77 76 75 74 73 72 11 70 69 68 67 66 65 64 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49 48 41 46
G--G&C-G-T--G 1 - ( 1 5 ) 1
1
HI
U G U G G C G C G U A G U
I 1 C-C-C-G-C-A-C-A-C-C-G-C
G G G
77 76 75 74 73 72 11 70 69 68 67 66 65 64 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49 48 41 46
C’
C
324
The Total Synthesis of Nucleic Acids
With this successful application of the ligase enzyme in joining synthetic polynucleotides, it was considered essential to study the fidelity of the T,-infected E. coli ligase enzyme. Recently it has been discovered122that this enzyme can also catalyze the covalent joining of interrupted deoxyribooligonucleotide strands with one mispaired base at the 3’-hydroxyl terminus of a bihelix (P3ZTllC Poly dA). Similar types of mismatched joining as well as end-to-end dimerizations have subsequently been reported to occur during the synthesis of alanine t RNA gene.lZ3 A recent report showed that terminal crosslinking in DNA strands by an enzyme system consists of the T,-ligase and another activity considered likely to be an e x o n ~ c l e a s e . ~ ~ ~
+
8.
CONCLUSION
Two distinct periods of nucleic acid chemistry have thus far emerged. In the pre-1960 period, some adventuresome organic chemists took up the challenges implicit in the structural complications of these macromolecules. The main impetus was provided by the Watson-Crick hypothesis of the early 1950s. In the period following 1960, as much of the basic chemistry was elaborated, more chemists and biologists began to contribute to this problem and increasingly complicated molecules were synthesized; this culminated in 1970, with the first announcement of the test-tube synthesis of a gene. After this achievement, which undoubtedly constitutes a milestone in the development of new experimental techniques, it is probably safe to predict another period of prolific advancement in the 1970s. What areas need additional efforts or will experience growth? Some of the synthetic reagents require additional sophistication, two of the more important being those for effecting the formation of the internucleotidic bond and for differentiating the 2’-hydroxyl group in the ribonucleotides. The former is required to make the condensation faster and quantitative when using stoichiometric amounts of the two components to be condensed. There is always a need for faster separation techniques to relieve some of the tedious parts of the synthetic work in this field. But more important is the necessity for devising easier techniques for the sequence determination of the longer synthetic deoxypolynucleotides. The discovery of new enzymes and their ready availability in the pure form will undoubtedly assist nucleic acid chemists. As techniques become better established, not only will other “genes” be synthesized but these will be incorporated into the natural DNA. In fact, at the rate this field is advancing such announcements may well be made before this review is published.
References
325
ACKNOWLEDGMENT
We are grateful to Mrs. Gloria Dumoulin for her patience in typing the manuscripts.
REFERENCES 1.
H. G. Khorana, H. Buchi, H. Ghosh, N. Gupta,T. M. Jacob, H. Kossel, R. Morgan, S . A. Narang, E. Ohtsuka, and R. D. Wells, Cold Spring Harbor Symp., 31, 39
2.
A. M. Michelson and A. R. Todd, J. Chem SOC.,1955, 2632. H. G. Khorana, in D. Shugar, Ed., Genetic Elements: Properlies and Furirtion. Academic Press, New York, 1966, p. 209. A. M. Michelson, The Chemis/ry of Nucelosides and Niirleotides, Academic Press, London, 1963. H. G. Khorana, Some Recent Developnrents in the Chemistry ofPhosphate Esters of Biological Interest, Wiley, New York. 1961. W. W. Zorbach and R. S . Tipson, Eds., Synthetic Proredtires in Niirleic Acid Chenristry, Interscience, New York, 1968. (a) F. Crarner, Pure Appl. Chem., 18, 197 (1969); (b) H. G. Khorana, Pure Appl. Chetn.. 17, 349 (1968); (c) F. Cramer, Angeiv. Cheni. h t . Ed., 5, 173 (1966). For typical examples, see (a) R. L. Letsinger and W. S . Mungall, J. Org. Chenr., 35, 3800 (1970); (b) G. H. Jones, H. P.Albrecht, N. P. Damodaran, and J. G. Moffatt, J . Ant. Chern. Soc., 92, 5511 (1970); (c) D. A. Shurnan, M. J. Robins, and R. K. Robins, J. Am. Chem. Soc., 92, 3434 (1970); (d) M. Ikehara, Accts. Cheni. Res., 2,
(1966). 3. 4.
5. 6. 7. 8.
47 (1969).
For typical examples, see (a) R. S. Goody, K. A. Watanabe. and J. J. Fox, Tetrahedron Lett., 1970, 293; (b) T. Y. Shen, Angew. Chenr. Int. Ed., 9 , 678 (1970); (c) D. Suhadolnik, Nuceloside Antibiofics, Wiley, New York. 1970. For typical examples, see (a) R. H. Hall, M. J. Robins, L. Stasiuk, and R. Thedford, 10. J. Am. Chem. Soc., 88, 2615 (1966); (b) H. G. Zachau, Angew. Chem. Itit. Ed., 8, 9.
711 (1969). 11. (a) D. Brown, in R. A. Raphael, E. C. Taylor, and H. Wynberg, Eds., Advances in Organic Chemistry-Methods and Resulls, Vol. 3, Interscience, New York, 1963, p. 75; (b) V. M. Clark, D. W. Hutchinson, A. J. Kirby, and S . G. Warren, AnReiv. Chern. Int. Ed., 3 , 678 (1964).
12. (a) K. Imai, S. Fujii, K. Takanohashi, Y.Furukawa. T. Masuda, and M. Honjo, J. Org. Chem., 34, 1547 (1969); (b) T. A. Khawaja and C.B. Reese, J . Am. Clzem. Soc., 88, 3446 (1966); (c) R. Saffhill, J. Org. Chern., 35, 2881 (1970). 13. F. Cramer and H. NeunhoetTer, Chem. Ber., 95, 1664 (1962). 14. R. von Tieerstrorn and M. Smith. Science. 167. 1266 (19701.
326
The Total Synthesis of Nucleic Acids
15. A. M. Michelson, J. Chem. SOC.,1959, 3655. 16. R. Lohrmann and H. G. Khorana, J. Am. Chem. SOC.,88, 829 (1966). 17. F. Cramer, R. Wittmann, K. Daneck, and G. Weimann, Angeiv. Chem. fnr. Ed., 2, 43 (1963). 18. M. Smith, J . G. MotTatt, and H. G. Khorana,J. Am. Chem. Soc., 80, 6204 (1958). 19. F. Cramer, H. Neunhoeffer, K.H. Scheit, G. Schneider, and J. Tennigkeit, Angew. Chem. fnt. Ed., 1, 331 (1962). 20. M. lkehara and H. Uno, Chem. Pharm. Bull. (Japan), 12, 742 (1964). 21. (a) F. Cramer, W. Rittersdorf, and W. Bohm, Jiistus Liebigs Ann. Chew., 654, 180 (1962); (b) F. Cramer, H. J. Baldauf, and H. Kuntzel, Angew. Chem. Int. Ed., 1, 54 (1 962). 22. T. M. Jacob and H. G. Khorana, J . Am. Chem. Soc., 86, 1630 (1964). 23. Y. Wolman. S. Kivity, and M. Frankel, Chem. Comm., 1967,629. 24. R. H. Ralph and H. G. Khorana. J. Am. Chem. Soc., 83, 2926 (1961). 25. (a) K. K. Ogilvie and R. L. Letsinger, J. Org. Chem., 32, 2365 (1967); (b) R. L. 91, 3356 (1969). Letsinger and P. S. Miller, J. Am. Chem. SOC., 26. (a) J. Zemlicka, S.Chlidek, A. Holy, and J. Smrt, Coil. Czech. Chem. Comm., 31, 3198 (1966); (b) A. H61y and J. Zemlicka, Coll. Czech. Chem. Comm., 34, 2449 (1969). 27. P.T. Gilham and H. G. Khorana, J . A m Chenr. Soc.. 80,6212 (1958). 28. M. Smith, D. H. Rammler, I. H. Goldberg, and H. G. Khorana, J. Am. Chem. Soc., 84, 430 (1962). 29. G.Weimann and H. G . Khorana, J. Am. Chenr. Soc., 84, 4329 (1962). 30. B. E. Griffin, M.Jarman, and C. B. Reese, Tetrahedron, 24,639 (1968). 31. C. B. Reese and J. C. M. Stewart. Tetrohedrorr Leu., 1968,4273. 32. R. K. Ralph and H. G. Khorana, J. Am. Chem. Soc., 83,2926 (1961). 33. H. G . Khorana and J. P. Vizsolyi, J . Am. Chem. SOC.,83, 675 (1961). 91, 3360 (1969). 34. R. L. Letsinger, K. K.Ogilvie, and P. S. Miller, J . Am. Chem. SOC., 35. R. L. Letsinger and V. Mahaderan, J. Am. Chem. Soc., 87, 3526 (1965). 36. A. F. Cook, J. Org. Chem., 33, 3589 (1968). 37. L. A. Cohen and J. A. Steele, J. Org. Chern., 31, 2331 (1966). 38. T. Hata and J. Azizian, Tetrahedron Lett., 1969, 4443. 39. A. F. Cook and D. T. Maichuk, J . Org. Chem., 35, 1940 (1970). 40. R. L. Letsinger and K. K. Ogilvie, J . Org. Chem., 32, 296 (1967). 41. (a) F. Cramer, H. P.Bar, H. J. Rhaese, W. Saenger, K. H. Scheit, and G. Schneider, Tetrahedron L e t / . ,1963, 1039; (b) S. Chlidek and J. Smrt, Coil. Czech. Chern. Comm. 29,214 (1964). 42. H. S. Sachdev and N. A. Starkovsky, Tetrahedron Lert., 1969, 733. 43. (a) Y. Lapidot and H. G. Khorana, J. Am. Chem. SOC.,85, 3852 (1963); (b) D. H. Rammler, Y.Lapidot, and H. G. Khorana. J . Am. Chem. Soc., 85, 1989 (1963). 44. (a) J. Smrt and F. Sorm, Coil. Czech. Chem. Comm., 28, 61 (1963); (b) D. H. Rammler and H. G. Khorana, J. Am. Chem. Soc., 84,3112 (1962). 45. J. Smrt and S. Chlidek, Coil. Czech. Chem. Comm., 31, 2978 (1966). 46. C. B. Reese, R. Saffhill, and J. E. Sulston, Tetrahedron, 26, 1023, 1031 (1970).
References
327
47. R. Lohrmann, D. Soll, H. Hayatsu, E. Ohtsuka, and H. G. Khorana,J. Am. Chem. SOC.,88, 819 (1966); see also references 30 and 44b. 48. A. Hampton and A. W. Nichol, Biochetn., 5,2076 (1966); see also reference 40. Liebigs Ann. Chem., 49. F. Cramer. W. Saenger, K. H.Scheit, and J. Tennigkeit, JUSIUS 679, 156 (1964); see also references 28,44a. 50. (a) J. Zemlicka and S. ChlBdek. Tetrahedron Le/f., 1965,3057; (b) C. B. Reese and J. E. Sulston, Proc. Chem. SOC.(London), 1964,214; (c) F.Eckstein and F. Cramer, Chem. Ber., 98, 995 (1965); (d) S. Chlhdek and J. Zemlicka, Coll. Czech. Chem. Comm., 32, 1776 (1967); see also reference 31. 51. G. M. Tener, J. Am. Chem. SOC.,83, 159 (1961). 52. D. So11 and H. G. Khorana, J. Am. Chem. SOC..87, 360 (1965). 53. W.Freist, R. Helbig. and F.Cramer, Chem. Ber., 103, 1032 (1970). 54. (a) F. Eckstein, Chem. Ber., 100, 2228 (1967); (b) A. Franke, F. Eckstein, K. H. Scheit, and F. Cramer, Chetn. Ber., 101, 944 (1968). 55. A. F. Cook, M. J. Holman, and A. L. Nussbaum, J. Am. Chenr. SOC.,91,1522,6479 (1969). 56. S. A. Narang, 0. S. Bhanot, J. Goodchild, S. K. Dheer, and R. H.Wightman, J. Chem. SOC.(D), 1970, 516. 57. F. Cramer and K. H. Scheit, Justus Liebigs Ann. Chem., 679,150 (1964).
58. F. Cramer,
H.P.BBr, H.J. Rhaese, W.Saenger. K. H. Scheit, and G. Schneider,
Tefrahedron Lett., 1963, 1039.
S. A. Narang, 0. S. Bhanot, J. Goodchild, and R. H. Wightman,J. Chem. SOC.(D), 1970,91; see also reference 82. 60. F. R. Atherton, H. T. Openshaw, and A. R. Todd, J. Chem. SOC.,1945,382. 61. K.H. Scheit, Tetrahedron L e / / . , 1967, 3243. 62. (a) E. Ohtsuka, K. Murao, M. Ubasawa, and M. Ikehara, J. Am. Chem. SOC.,91, 1537 (1969); (b) E. Ohtsuka, M. Ubasawa, and M. Ikehara, J. Am. Chem. SOC.,92, 59.
3441, 5507 (1970). 63. E. Ohtsuka, M. Ubasawa, and M. Ikehara, J. Am. Chem. SOC.,92, 3445 (1970). 64. F. Kathawala and F. Cramer, Justus Liebigs Ann. Chem., 712, 195 (1968). 65. (a) G. M. Ttner, H. G. Khorana, R. Markham. and E. H. Pol, J. Am. Chem. SOC., 80,6223 (1958); (b) A. F. Turner and H. G. Khorana, J. Am. Chem. SOC.,81, 4651 (1959). 66. H. G . Khorana and J. P.Vizsolyi, J. Am. Chem. SOC.,83,675 (1961). 67. (a) R. K. Ralph and H. G. Khorana, J. Am. Chem. Soc., 83,2926 (1961); (b) H.G. Khorana, A. F. Turner, and J. P. Vizsolyi. J. Am. Chem. Soc., 83, 686 (1961); (c) R. K. Ralph, W.J. Connors, H. Schaller, and H. G. Khorana, J. Am. Chem. Soc., 85, 1983 (1963). 68. H. G. Khorana, J. P. Vizsolyi, and R. K. Ralph, J. Am. Chem. Soc.. 84,414 (1962). 69. F. N. Hayes and E. Hansbury, J. Am. Chem. Soc., 86,4172 (1964). 70. 0. Pongs and P. 0. P. Ts’o, Biochem. Biophys. Res. Comm.. 36,475 (1969). 71.
S. A. Narang. 0. S. Bhanot, S. K. Dheer, J. Goodchild, and J. J. Michniewicz.
72.
E. Ohtsuka, M. W. Moon, and H. G. Khorana, J. Am. Chem. SOC.,87, 2956
Biochetn. Biophys. Res. Comtn., 41, 1248 (1970).
(1965).
328 73. 74. 75. 76.
The Total Synthesis of Nucleic Aclds S . A. Narang. T. M. Jacob, and H. G . Khorana, J. Ant. Chem. Sor., 89,2167 (1967). J. M. Jacob, S. A. Narang, and H. G. Khorana,J. A m . Chem. Soc., 89,2177 (1967). S. A. Narang and H. G. Khorana, J. A m . Chem. Soc., 87, 2981 (1965). For a leading reference, see E. Ohtsuka and H. G. Khorana, J . Am. Chem. Soc., 89,
2195 (1967). 77. T. M. Jacob and H. G. Khorana, J . A m . Cheni. Soc., 87, 368 (1965). 78. T. M. Jacob and H. G. Khorana, J. A m . Chem. Soc., 87,2971 (1965). 79. K . L. Agarwal, H. Biichi, M. H. Caruthers, N. Gupta, H.G. Khorana, 80. 81. 82. 83. 84. 85. 86.
K.Kleppe, A. Kumar, E. Ohtsuka, U. L. Rajbhandary, J. H. Van de Sande, V. Sgaramella, H.Weber, and T. Yamada, Nature, 227, 27 (1970). R. L. Letsinger and K.K. Ogilvie, J. A m . Chent. Soc., 91, 3350 (1969). T. Neilson, J. Chem. SOC.( D ) , 1968, 1139; see also reference 54. C. B. Reese and R. Saffhill, Chem. Conim., 1968, 767. S. A. Narang, T. M. Jacob, and H. G . Khorana,J. A m . Chem. Soc., 89,2158 (1967). S. A. Narang and S. K. Dheer, Biochem., 8, 3443 (1969). J. J. Michniewicz, 0. S. Bhanot, S. K. Dheer, J. Goodchild, R. Wightman, and S. A. Narang. Biochim. Biophys. Acta, 224, 626 (1970). (a) H. Hayatsu and H. G. Khorana, J. Ani. Chern. SOC.,89, 3880 (1967); (b) F. Cramer, R. Helbig, H. Hettler, K. H. Scheit, and H. Seliger, Angew. Chern. Int. Ed., 5 , 601 (1966).
87. 88. 89.
90. 91. 92. 93. 94.
(a) R. L. Letsinger and V. Mahadevan, J . Am. Chem. Soc., 87, 3256 (1965); (b) L. R. Melby and D. R.Strobach, J. A m . Chem. SOC.,89, 450 (1967). L. R. Melby and D. R. Strobach. J. Org. Cherti., 34, 421, 427 (1969); see also reference 86. (a) R. L. Letsinger, M. H.Caruthers, and D. M. Jerina, Biochem., 6 , 1379 (1967); (b) T. Kusama and H. Hayatsu, Chem. Pharm. Bull. (Japan), 18, 319 (1960); (c) T. S. Shmidzu and R. L. Letsinger, J . Org. Chem., 33, 708 (1968). R. L. Letsinger and V. Mahadevan, J . Am. Chem. Soc., 88, 5319 (1966). G. M. Blackburn, M. J. Brown, and M. R. Harris, J . Chern. Soc. (C), 1967, 2438. F. Cramer and H. KBster, Augew. Chem. h i t . Ed., 7, 473 (1968). C. Coutsogeorgopoulus and H. G. Khorana, J. Am. Cheni. Soc., 86, 2926 (1964). For typical examples, see (a) K. K. Ogilvie and D. Iwacha, Can. J. Clreni., 48, 862 (1970); (b) J. Nagyvary, Biocheni., 5, 1316 (1966); (c) Y.Mizuino and T. Sasaki, Tefrahedron Lett., 1965,4579; (d) P. C. Srivastava, K. L. Nagpal, and M. M.Dhar, Experentia, 25, 356 (1969).
95.
For typical examples, see (a) H.J. Rhaese, W. Siehr, and F. Cramer, Justtts Liebigs A m Chem.. 703, 215 (1967); (b) B. E. Griffin and C. B. Reese, Tetrahedron, 24, 2537 (1968); (c) A. Holi and J. Smrt, Coll. Czech. Chem. Comm., 31, 3800 (1966);
see also references 30, 43-50, and 63. (a) A. Kornberg, I. R. Lehman, M. J. Bessman, and E. S. Simms, Biochim. Biophys. Acta, 21, 197 (1956); (b) I. R. Lehman. M. J. Bessman, E. S. Simms, and A. Kornberg, J . Biol. Cheni., 233, 163 (1958); (c) C. C. Richardson, C. L. Schildkraut, H. V. Aposhian, and A. Kornberg, J. Biol. Chem., 239,222 (1964). 97. A. Kornberg. L. L. Bertsch, J. F. Jackson, and H. G. Khorana, Proc. Nat. Acad. Sci. U . S . , 51, 315 (1964).
96.
References
329
98. (a) C. Byrd, E. Ohtsuka, M. W. Moon, and H.G. Khorana, Proc. Nut. Acud. Sci. U.S., 53, 79 (1965); (b) R. D. Wells, E. Ohtsuka, and H. G. Khorana, J. Mol. Biol.. 14,221 (1965). 99. R. D.Wells, T. M. Jacob, S. A. Narang, and H. G. Khorana, J. Mol. Biol., 27, 237 (1967). 100. R. D. Wells, H. Biichi, H. Kossel, E. Ohtsuka, and H. G . Khorana, J. Mol. Biol., 27, 265 (1967). 101. H. Schachmann, J. Adler, C. Radding, I. Lehman, and A. Kornberg, J . Biol. Cliem., 235, 3242 (1960). 102. (a) F. J. Bollum, E. Groeninger, and M. Yoneda, Proc. Nat. Acud. Sci. US.,51, 853 (1964);(b) M. Yoneda and F. J. Bollum, J. Biol. Chenr., 240, 3385 (1965). 103. F.Bollurn, Fed. Proc., 24 (No. 2, Part I), 1207 (1965). 104. (a) J. Hurwitz, J. August, J. Davidson, and W. E. Cohen, Progress in Nrrcleic Acid Research, Vol. I, Academic Press, New York, London, 1963,p. 59; (b) J. Hurwitz, Methods in Enzymology, Vol. VI, Academic Press, New York,London, 1963,p. 23; (c) M. Chamberlin and P. Berg, Proc. Nat. Acud. Sci. US.,48, 81 (1963); (d) E. Fuchs, W.Zillig, P. H. Hofschneider, and A. Preuss, J. Mol. Biol. 10, 546 (1964); (e) M. Chamberlin and P. Berg, J. Mol. Biol., 8,297 (1964). 105. S . Nishimura, T. M. Jacob, and H. G. Khorana, Proc. Nut. Acud. Sci. U.S.,52, 1492 (1964). 106. S. Ochoa, Angew. Clreni., 72, 225 (1960). 107. S. P. Colowick and N. 0. Kaplan, Methods in Enzymology, Vol. VI, Academic Press, London, New York, 1962,p. 27. 108. M. F. Singer, R. J. Hilmoe, and M. Greenberg, J . Bid. Chrni., 235, 2705 (1960). 109. F. Cramer, H. Kuntzel, and J. H . Matthaei, Atgeiv. Clrmi. f n t . Ed., 2, 589 (1964). 110. R. E. Thach and P. Doty, Science, 184,632 (1965). 111. R. W. Holley, J. Apgar, G. A. Everett, J. T. Madison, M. Marquise, S.H. Merrill, J. R. Penswick, and A. Zaniir, Science, 147, 1462 (1965). 112. F.Egami and K. Sato-Asaon, Biochem. Biopliys. Acta, 29, 655 (1958). 113. (a) C. C. Richardson,Proc Nut. Acud. Sci. U.S., 54,158(1965);(b) A. Novogrodsky, M. Tal, A. Traub, and J. Hurwitz, J. Biol. Chenr., 241, 2933 (1966). 114. A. Garen and C. Levinthal, Biochitn. Biophys. Actu, 38, 470 (1960). 115. W. E. Razzell and H. G. Khorana, J. Biol. Clienr., 234,2105 (1959). 116. W. E. Razzell and H. G . Khorana, J . Biol. Chetn., 236, 1144 (1961). 117. (a) S. B. Zimmerman, J. W. Liltle. 0. K. Oshinsky, and M. Gellert, Proc. Nut. Acud. Sci. U.S.,57, 1841 (1967); (b) B. Olivera and I. R. Lehman, Proc. Nat. Acud. Sci. U.S., 57, 1426 (1967); (c) M. L. Gefter, A. Becker, and J. Hurwitz, Proc. Nat. Acud. Sci. US.,58,240 (1967);(d) B. Weiss and C. C. Richardson, Proc. Nut. Acud. Sci. US.,57, 1021 (1967);(e) N. R. Cozzarelli, N. E. Melechen, T. M. Jovin. and A. Kornberg, Biocheni. Biopliys. Res. Comitiiin., 28, 578 (1967). 118. B. M. Olivera, Z. W. Hall, Y. Anraku, J. R. Chien, and 1. R. Lehman, Cold Spring Harbor Symp., 33,27 (1968). 119. N . K. Gupta, E. Ohtsuka, H.Weber, S. H. Chang, and H. G. Khorana, Proc. Nut. Acud. Sci. US.,60,285 (1968).
330
The Total Synthesis of Nucleic Acids
120. The New York Times, CXIX, 1 (June 3, 1970). 121. N. K . Gupta and H. G . Khorana, Proc. Nar. Acad. Sci. U.S., 61,215 (1968). 122. C. M. Tsiapalis and S. A. Narang, Biochem. Eiophys. Res. Commun., 39,631 (1970). 123. V. Sgaramella, J. H. van de Sande, and H. G. Khorana, Proc. Nar. Acad. Sci. US., 67, 1468 (1970). 124. B. Weiss, Proc. Nut. Acad. Sci. US.,65, 652 (1970).
Total Synthesis OfNatural Products, Volume1 Edited by John Apsimon Copyright © 1973 by John Wiley & Sons, Inc.
The Total Synthesis of Antibiotics FRANCIS JOHNSON The Dow Chemical Company Eastern Research Laboratory Waytand, Massachuseffs
1. Introduction
2. The Penicillin Group 3.
4.
5.
6.
7.
A. The Penicillins B. The Cephalosporins The Tetracyclines A. 6.Deoxy-6-Dernethyltetracyc~ine B. Tetracycline C. Oxytetracycline The Basic Sugars A. The Kanamycins B. Kasugamycin C. Streptozotocin D. Lincomycin Nucleoside Antibiotics A. Purine Nucleosides B. Pyrrolopyrimidine Nucleosides C. Miscellaneous Nucleosides Peptide and Depsipeptide Antibiotics A. Tyrothricin B. The Polymixins C. Cyclic Depsipeptides Macrolide Antibiotics A. Curvularin
332 331 338 342 348 349 356 359 364 361 377 379 380 381 390 398 400 404 406 413 419 426 428
331
332 8.
The Total Synthesis of Antibiotics B. Zearalenone Miscellaneous Antibiotics
A. Anthramycin B. Griseofulvin C. Novobiocin D. Cycloheximide E. Mitoniycin C F. Chloramphenicol References
429
434
437 440 447 451 455 457 458
1. INTRODUCTION
Organic substances that can inhibit or destroy one form of life or another are extremely common in nature. Their sources are found in almost all but the highest life forms, ranging from the lowly bacterium, such as Clostridium botulinum, which produces one of the most deadly toxins known to man, through the plant and insect kingdoms to the lower forms of aquatic and terrestrial life. In the latter categories one can mention the puffer fish, the eggs and ovaries of which contain one of the most potent toxic substances known-tetrodotoxin (LD,, in mice: 10 pg/kg, given ip.)’”-and Russell’s viper, which harbors one of the deadliest venoms yet found [LD,, in mice: 0.13 mg/kg given iv.].’” In many instances the real biological reasons for the existence of such substances are not known. However, it is obvious that frequently toxic substances are used defensively for self-protection and the protection of territory against predators and aggressors. Conversely, there are many instances where predators use such materials aggressively to render their victim harmless. In fact they constitute nature’s arsenal for waging chemical warfare and taken in perspective this appears to be one of the commonest methods employed among the various species in their fight for continued existence. Antibiotics are specific chemical compounds derived from or produced by living organisms that can, in small amounts, selectively inhibit the life processes of other organisms. In general, however, the term “antibiotic” frequently refers only to substances that inhibit microorganisms which in particular are pathogenic to man, other higher animals, or plants. Such microorganisms include bacteria, mycobacteria fungi, amebae, and more rarely viruses. For an antibiotic to be useful therapeutically it should have a high order of selective toxicity, that is, it should have little or no toxicity to the disease-bearing host at those levels at which it inhibits or kills the disease-producing organisms. Predicated on this requirement an antibiotic should be capable of being absorbed by the oral route (except where topical application is superior) and should remain at adequate levels in the blood
1. Introduction
333
for a reasonable period of time. Axiomatically perhaps, it should achieve wide distribution throughout the various tissues of the body. Faced with such limitations very few organic compounds qualify as chemotherapeutic agents. Those that do largely are of microbial origin, although a few come from higher forms of life, a few are now made synthetically, and a number are actually hybrids of biochemical and laboratory syntheses. The latter are essentially laboratory modifications of existing antibiotics and such modified compounds are becoming increasingly necessary as new strains of microorganisms resistant to the parent antibiotic make their appearance. For example, in the cases of the penicillin and tetracycline antibiotics, substantially superior derivatives have been prepared by synthetic modification of the basic molecules. Again in the case of the rifamycins minor changes in structure led to a derivative that was not only less toxic to the host but also achieved much higher blood levels. Even here, however, there appears to be a limit because there is some evidence, albeit tenuous, that synthetic modification is only worthwhile when the microorganism(s) producing the antibiotic, is itself capable of producing a range of structural variants. When the substance produced is unique in structure, synthetic modification has failed, essentially, to produce a more effective antibiotic (e.g., in the cases of griseofulvin, fusidic acid, and chloramphenicol). In this essay we are concerned only with the total synthesis of those antibiotics that are therapeutically useful (or whose derivatives are useful) and that are of microbiological origin. Also included are a number of drugs that look promising but are still under investigation. Table 1 lists alphabetically (within group classifications) the more important of the useful antibiotics of known structure, their microbial orgin, and their uses and indicates the status of the synthetic art surrounding them. A limited number of partial syntheses have also been documented where the chemistry is significant. Antibiotics (e.g., emetine, the vinca alkaloids, and camptothecin) derived from higher plant life are not discussed, nor are antibiotics of nonbiogenetic origin such as the sulfonamides, the nitrofurans, or nalidixic acid. Before proceeding to a discussion of the total syntheses accomplished so far it must be stated that only in rare cases has a totally synthetic product succeeded commercially, in supplanting the biosyntheticallyproducedmaterial. Generally this only happens when the molecule under consideration is extremely simple [such as aquamycin (NH,COC=CCONH,) used against rice blast] and can be produced in a few highly efficient and inexpensive steps. By comparison with life processes, laboratory methods are both clumsy and wasteful and rarely can be justified by the economics involved. Why then do chemists attempt the synthesis of what are for the most part highly complex and often unstable molecules? Part of the answer lies primarily
Table 1. Therapeutically Useful Antibiotics of Known Structure
Name Basic Sugars Gentamicin Kanamycin Kasugamycin Lincomycin Neomycin-B Paromomycin
Spectinomycin Streptomycin Streptozotocin
Source
Gram-positive and negative bacteria Gram-positive and negative bacteria; mycobacteria; leptospira Streptomyces Pyricularia oryzae (rice kasugaensis blast pathogen) Streptornyces Gram-positive bacteria; lincolnensis m ycobacteria Streptornyces fradiae Gram-positive and negative bacteria; mycobacteria Streptomyces rimosus Gram-positive and negative J paromomycicus bacteria; mycobacteria; amebic infection and murine leprosy Streptomyces Gram-positive and negative spectabilis bacteria Streptomyces griseits Gram-positive and negative bacteria; mycobacteria Streptomyces Gram-positive and negative achromogenes bacteria; tumor cells Micromonospora purpitrea Slreptomyces kanamyceticus
Macrolides Amphotericin-B Streptomyces nodosus Carbomycin Streptomyces lialstedii Chalcomycin Erythromycin
Filipin Nys t at in
334
Used Against
Streptomyces bikiniensis Streptomyces erythreus Streptomyces jlipinensis Streptomyces albulirs
Yeasts and fungi Gram-positive bacteria rickettsias; protozoa; penicillin-resistant staphylococci Gram-positive and negative bacteria; mycobacteria Gram-positive bacteria; rickettsias; protozoa; penicillin-resistant staphy loccoci Yeasts and fungi Yeasts and fungi (mainly topical use)
Total Synthesis
Yes Yes
Yes Yes
No No
No
No YeS
No
No No No
No No
Table 1. (continued) Name
Source
Oleandomycin
Streptomyces antibioticus
Spiramycin
Streptomyces ambofaciens Streptom-ycesfradiae
Tylosin
Nucleosides Angustmycin-A Streptomyces hygroscopicus Blast icidi n-S Strep?omyces griseoehromogenes Nucleocidin Streptomyces calvus Streptomyces cacaoi var. asoensis Streptomyces Psicofuranine hygroscopicus Streptomyces Puromycin alboniger Racemomycins- Streptomyces racemochromogenus A and 0 Unidentified Sangivamycin streptomyces Streptomyces Septicidin fimbriatus Strep tomyces Toyocamycin toyocaensis Streptomyces Tubercidin tubercidicirs Penicillin group Cephalosporium Cephalosporin-C species
Polyoxins
Penicillin
Penicillium notatum
Polypeptides and depsipeptides Actinomycin-D Streptomyces antibioticus
Used Against Gram-positive bacteria; rickettsias; protozoa; penicillin-resistant staphylococci Gram-positive bacteria; rickettsias Gram-positive bacteria including mycobacteria
Total Synthesis
No
No
No
Primarily mycobacteria
Yes
Piricularia oryzal
No
Gram-positive and negative bacteria; tripanosomes Piricularia oryzal
No
No
Tumor cells
YeS
Gram-positive bacteria ; tumor cells; protozoa Gram-positive and negative bacteria; mycobacteria Leukemia cells
Yes No
Tumor cells; fungi
No
Tumor cells
Yes
Tumor cells; mycobacteria
Yes
Penicillin-resistant organisms, especially staphylococci Gram-positive bacteria; spirochetes ; actinomycetes
Yes
Tumors
Yes
Yes
Yes 335
Table 1. (continued)
Name
Source
Bacitracin
Bacillrrs srrbtilis
Gramicidin
Bncillris brevis
Polymixin-B,
Bucillirs yolyttiixa
Stap hy loinyci n
Streprotnyces fradiae
Tyrocidine
Bacillus brevis
Tetracycline group Streptotnyces Chlorotetracycline arrr.eofacieris OxytetraSpr reptoinyces ritnosrrs cycline Miscellaneous Ant hrani yci n Chloraniphenicol Cycloserine Cournermycin
Streptotnyces refirincus Streptornyces vetiezirelae Sprrepprotnyces orchidacerrs Streyproniyccs rishiricnsis
Cyclohexiinide
Streptotnyces griseris
Daiinomycin
Streprorriyces perrcetirrs Aspergillirs ,@iigarirs Frrsideirin coccineiirn
Fumagillin Fusidic acid Griseofulvin Mitoniycin-C 336
Penicilliritii griseofirluiii Streptoniyces caespitosrrs
Used Against
Total Synthesis
Gram-positive bacteria; amebae (mainly used topically) Gram-positive bacteria (mainly topical application) Gram-negat ive bacilli except proteus Gram-positive and mycobacteria Gram-positive bacteria (mainly topical application)
No
Gram-positive and negative rickettsias; large viruses Gram-positive and negative bacteria; rickettsias; large viruses
No Yes
Tumor cells
Yes
Gram-positive and negative bacteria; rickettsias; large viruses Mycobacteria
Yes
Yes Yes No
Yes
Yes
Gram-positive bacteria including mycobacteria ; gram-negative bacteria Phytopathic fungi; tumor cells Tumor cells
No
Amebae Gram-positive bacteria; mycobacteria Mycotic infections of hair, nails and skin Tumor cells
No
Yes No No
Yes No
2. The Penicillin Group
337
Table 1. (Continued)
Name Monensin Novobiocin
Rifamycin-B Streptonigrin Streptovitacin-A
Source
Used Against
Antibacterial (under investigation) Streptomyces niverrs Gram-positive and negative bacteria ; penicillinresistant staphylococci Streptomyces Primarily mycobacteria; mediterranei gram-positive bacteria Streptomycesjocculus Tumor cells Streptomyces Fungi; protozoa; tumor griserrs cells Streptomyces cinnamonensis
Total Synthesis No Yes
No
No No
in the challenge itself and secondarily in the attendant satisfaction and status that comes with success-the same reasons that led Hillary to attempt the scaling of Everest. Total synthesis is one of the highest aims a chemist can aspire to. As Woodwarda has said: “There is excitement, adventure and challenge and there can be great art in organic synthesis.” Apart from the human side, very few syntheses of complex molecules are accomplished without the concomitant development of either new ideas and/or new synthetic methods. These reasons alone are sufficient justification. In a few instances total syntheses are undertaken to provide a structure proof for a compound, but now that spectroscopic methods are so highly developed authentic cases of this are rare. In the discussion that follows the antibiotics are grouped where possible, as they are in Table I , according to structural similarity. However, the first to be considered is the penicillin group not only because of its importance but because of the prime position penicillin itself occupies in the history of antibiotics. 2. THE PENICILLIN GROUP
The term “penicillin” usually refers to compounds that have the general structure represented by 1 in which R is any acyl group. In the naturally occurring substances this acyl function is phenylacetyl (benzyl penicillin) or 5-carboxy-5-aminovaleroyl (penicillin N). Variations in the acyl function can be achieved in a limited number of cases by the addition of appropriate alcohols or acids to the fermentation medium. However, the development of both synthetic and biochemical methods3 of preparing 6-aminopenicillanic
338
The Total Synthesis of Antibiotics
11 H
CH, CH,OCOCH,
C0,H
CO,H 1
2
acid (1; R = H) has permitted access to a wide variety of penicillins with differing acyl groups. From a therapeutic standpoint this has been of immense value since many of these new compounds are much move effective because of their resistance to cleavage by penicillinase. The latter is an enzyme, generated by some microorganisms, which inactivates the simpler penicillins by cleavage to 1 (R = H),which is itself only weakly antibiotic. The newer compounds' include methicillin (R = 2,6-dimethoxybenzoyl), ampicillin, (R = a-aminophenylacetyl), and nafcillin, (R = a-ethoxynaphthoyl), all of which are highly resistant to penicillinase. This is also true of the cephalosporins (2), the dominant naturally occurring form of which has R = D-5-amino-5-carboxyvaleroyl.The preferred' semisynthetic compound, cephaloridine, is the thienylacetyl derivative (2; R = [C4H3S]CHaCO)of 7-aminocephalosporanic acid 2 (R = H).
A. The Penicillins During World War I1 a prodigious amount of research5 was devoted to solving the structure of penicillin and thereafter to attempting its synthesis. The major difficulty in the synthesis of penicillin lies in the fragility of the /?-lactam. It has proved difficult to close such a ring and when closed, it is very easily cleaved again. In the cephalosporins the /?-lactam is not quite so sensitive because the ring strain associated with it is not so great, by reason of its being attached to a six- rather than a five-membered ring. The early wartime synthesis saw the preparation of benzyl penicillin (12) as its triethylammonium salt, in only minute yield.'" The complete synthesis shown in Scheme 1 involves as the critical step the condensation of Dpenicillamine 3 with 2-benzyl-4-methoxymethylene-5(4)oxazolone 4. The simplest and most practical synthesis that was devised8for the preparation of 3 involved the conversion of isobutyraldehyde to valine, by the Strecker reaction (60 % yield) followed by chloroacetylation to give 5 in 86 % yield. Treatment of 5 with acetic anhydride afforded the oxazolone 6 and the latter compound when treated with hydrogen sulfide in rnethanolic sodium methoxide yielded N-acetylpenicillamine (7). Acid hydrolysis then produced 3, which was resolved as its N-formyl derivative by means of brucine.
2. The Penicillin Group
80 %
(CH3),CHCH(CO2H)NNC0CHzCl 5
339
CH3 6
(CH&C(SH)CH(CO,H)NHCOCH,
%
7
(CH3),C(SH)CH(NH,)CO,H 3
C6H6CH2CONHCH2C0.$H3 % C6H6CH~CONHCH(CHO)C0,CH3 -%9
8
68 %
C,H,CH,CONHCHCO,H +
I
CWCHJ,
10
e 11 12
Scheme 1
The oxazolone 4 was preparedDby the acetic anhydride-induced cyclization of 10 itself obtained from 8 in three steps involving first formylation to give 9 followed then by ketalization and basic hydrolysis. The condensation of 3 as its hydrochloride with 4 was carried out in pyridine in the presence of triethylamine at -0' and the reaction mixture was partitioned between chloroform and a phosphate buffer, after 10 minutes. The chloroform-soluble fraction was again subjected to treatment with a pyridinetriethylamine solution containing a small amount of pyridine hydrochloride and heated for 7 minutes at 130". This was followed by a second phosphate
14
13
I
Oh' 15
16
17
HYcH33
C,H,OCH,CONHCH+ HO,CI HN
5%
"".CO H
18
19 Scheme 2
340
2. The Penicillin Group
341
buffer treatment after the pyridine-triethylamine mixture had been largely removed at low temperatures. Bioassay of the resulting material indicated a yield of 4 . 0 7 %. By extensive countercurrent partitioning and subsequent crystallization a pure sample (3.9mg) of the triethylammonium salt of 12 was obtained. Its physical properties, including the infrared spectrum, were identical with those of the salt obtained from natural benzyl penicillin. This synthetic approach was based on the thought that the base-catalyzed condensation would give an intermediate such as 11 which under the further influence of acid would cyclize to 12 to some degree, Other work had shown1o that 12, when treated with anhydrous HCI, gives the hydrochloride of 11. Undoubtedly an equilibrium is involved that greatly favors 11. In 1957 a rational synthesis (depicted in Scheme 2) of penicillin V (19) was announced by Sheehan and Henery-Logan." In their synthesis points worthy of note are (a) intermediates were used that would avoid azlactone formation and (b) again the construction of the /I-lactam was delayed until the last step. Formylation12 of the phthalimide 13 afforded 14, which, when treated with D-penicillamine in aqueous ethanol, yielded a mixture of two isomers (15), the more soluble of which (the a-isomer) corresponded to the natural product, in stereochemistry. When this isomer was split by hydrazine, 16 resulted. This was converted consecutively to 17 and then 18 by means of phenoxyacetyl chloride and anhydrous HCl, respectively. In the key step cyclization of 18 to penicillin V (19) was cleverly accomplished on the sodium salt in dilute aqueous dioxane by means of N,N-dicyclohexylcarbodiimide. This synthesis was quickly followed by a general synthesis of penicillin^'^ in which the key compound, 6-aminopenicillanic acid (23), was obtained as shown in Scheme 3. The amino group of 20 was first protected by means of a trityl function and the product, 21, was cyclized to the @-lactam22, using N,N-diisopropylcarbodiimide. Saponification of 22 afforded the corresponding carboxylic acid (I7 % yield) and detritylation was accomplished in 32% yield by means of dilute hydrochloric acid. The product was identical with 6-aminopenicillanic acid (23), a compound that can be acylated to produce any desired penicillin as noted previously. Recently an ingenious attempt14 has been made to construct the #?-lactam ring at an even earlier stage in the synthesis (Scheme 4). This involved the addition of azidoketene (from azidoacetyl chloride and triethylamine) to the thiazoline 24. Unfortunately not only was the yield of product low but the compound proved to be the 6-epi-isomer 25. Catalytic reduction of 25 gave 26 in moderate yield, which when acylated with phenoxyacetyl chloride afforded 6-epipenicillin V methyl ester (26a). A number of other approaches to the penicillin nucleus have been rebut as yet no new total synthesis has emerged.
342
The Total Synthesis of Antibiotics
20
21
13
22
Scheme 3
26 R = H 26a R = C6H60CH,C0
Scbeme 4
B. The Cephalosporins The total synthesis of cephalosporin C 41 has been accomplished directly by Woodward and his associates1' and the successful conversion of the penicillin nucleus to the cephalosporin nucleus has been reported by the Lilly group.18-20 In the total synthesis of cephalosporin the major difficulty that had to be overcome was the ease with which the cephalosporin analogs of penicillanic acid decarboxylate. The problem was avoided by using methods that avoided the generation of the free acid group. A critical intermediate was the /?-lactam
2. The Penicillin Group
343
33 and for its preparation (Scheme 5 ) the starting point chosen was the C02CH3
I
29
28 R - H 28a R = CH,
27
BdOCO-N s CH3%CH3
-
31
30
BU'OCON
H
s
H
CH3 CH, 32
Scheme 5
33
thiazolidine 27, previously prepared2' from L-( +)-cysteine. Treatment of 27 with r-butyloxycarbonyl chloride to protect the nitrogen atom afforded 28, whose methyl ester 28a underwent attack at the CH, group by dimethyl azodicarboxylate to give 29. Oxidation of 29 by means of lead tetraacetate followed by treatment of the reaction mixture for 24 hours with sodium acetate under anhydrous conditions led to the trans-hydroxyester30 (R = H). The corresponding tosylate (30; R = C,H,SO,) prepared in situ afforded the cis-azide 31 when treated with azide ion in water. Reduction of 31 at - 15" by alunhum amalgam gave the cis-amino ester 32, which in a novel way was converted, by means of triisobutylaluminum in toluene, to the /3-lactam 33. The structures of both 32 and 33 were confirmed by X-ray crystallography at this stage. In a parallel series of experiments (Scheme 6) di-/3,/3,/3-trichloroethyl d-tartrate was oxidized by sodium metaperiodate to the glyoxylate hydrate
C02CH,CC13
-p CO,CH,CCI,
C0,CH2CCI, CH(OH), I
I
+
+
C
/ \
O///&p
34
I I
CH
OHC
CHO
36
35
CO2CHZCC13
:qkR C02CHZCC13
=
C I 3CC H2 0 ZCC H (CHt ),CO N H
I N HCOzCH2CCl3
39 R = C H O 39a R = CH,OCOCH,
0
CO,CH,CCI,
byH2ococH3
'i
C13CCH20,C H(CH3)3CONH' NHCO2CH2CCI3 40
NH2
0b y H 2 0 c 0 c H 3
I
H O,CC(CHZ )&O NH 41 Scheme 6
344
4
2. The Penicillin Group
345
34. The latter when condensed with sodium malondialdehyde afforded 35,
which when added to distilling octane underwent dehydration giving the highly reactive dialdehyde 36. Condensation of 36 with the p-lactam 33 in n-octane at 80" led to the adduct 37, which in trifluoroacetic acid was smoothly converted in a masterful step, to the aminoaldehyde 38. Acylation of 38 to give 39 was accomplished using N-B,p,p-trichloroethylethoxycarbonyl-o-(-)-ar-aminoadipic acid in the presence of dicyclohexylcarbodiimide. Reduction in tetrahydrofuran by diborane followed by acetylation then led to the iso-ester 39a, which was equilibrated with the normal ester 40 by means of pyridine at room temperature, (Knorme,/Kiso= a). Reduction of 40 by zinc dust in acetic acid then removed all three trichloroethyl groups to give cephalosporin C (41), thus completing what must be regarded as a brilliant synthesis. In a separate series of experiments cephaloridine (cephalothin) (2; R = [C,H3S]CHzCO) was also synthesized. The partial synthesis of cephalosporin V (49; R = H) (Scheme 7) from penicillin V involves as the key step1* the rearrangement of the penicillin sulfoxide methyl ester 42 in boiling acetic anhydride. This gives rise, undoubtedly via the intermediate 43, to two products, the major one of which was assigned structure 45 and the minor one 44. Treatment of 44 with mild base afforded the desacetoxycephalosporin 46 (R = CH,), which also could be obtained more directly by heating 42 with a trace of acid in an inert solvent. For the purposes of completing19 the conversion, 46 (R = CH,) was hydrolyzed and converted to the p-methoxybenzyl ester (46; R = CH,OC,H,CH,). The latter was brominated by N-bromosuccinimide and the crude product treated with potassium acetate to give 47 (R = CH,OC,H,CH,), a mixture of the Az-and A3-isomers. Oxidation of this mixture with rn-chloroperbenzoic acid smoothly gave only the A3-sulfoxide 48 (R = CH,OC,H,CH,), thus providing a method of converting all of the cephalosporin material present to the biologically active A3-isomer. Reduction of 48 with sodium dithionite-acetyl chloride in DMF led to 49 (R = CH,OC,H,CH,-), which when cleaved by trifluoroacetic acid afforded cephalosporin V (49; R = H). In a second approach Spryz0prepared the sulfoxide 50 (R = N-phthalimidyl) by oxidation of 45 (R = N-pththalimidyl). Ring expansion of 50 by means of acetic anhydride and an acid catalyst gave a mixture of 51 (R = Nphthalimidyl) and 52 (R = N-phthalimidyl) in 30% yield. Conversion of 52 to 51 could be accomplished by acid-catalyzed dehydration. By using the p-nitrobenzyl ester in this sequence to protect the carboxyl group the conversion of 6-aminopenicillanic acid to 7-aminocephalosporanic acid was also achieved.
AcO-
42
43
0
CO,CH 44
CH,OAc
C 6 H 5 0 C H 2 c 0 N H ~ ~ c H ,
%// %//
0
CO,CH,
45
46
+ CH20Ac CO, R 41 Scheme 7
346
2. The Penicillin Group
347
0
48
49
CO, R
CO, R
Scheme 7 (Continued)
An interesting approach to the cephalosporin nucleus has been published by French workers22 and although it has not led to thesynthesis of a naturally 0
H H
C02CH3 51
0
H CO2CHS 52
occurring compound, it is worthy of inclusion. It is predicated on the work of Sheehan and is outlined in Scheme 8. The butenolide2s 53 was converted to 54 by means of thiolacetic acid and 54 then was hydrolyzed under acid conditions to the unstable thiol 55. Reaction of 55 with the phthalimido derivative 56 (itself prepared by the action of ammonium acetate on the corresponding formyl derivative) afforded the dihydrothiozine 57. Removal of the phthaloyl group followed by tritylation of the amine led to 58. The carboxyl group was then liberated
348
HO
x;;r
The Total Synthesis of Antibiotics
CHpN(CH,)pHCI
0
HO 0
CH,SAc
:%CHzSH
54
53
55
57
56
co,n d I
""""'9 0
0
0
58 R = trityl
59
59a
R = trityl R = thienylacetyl
Scheme 8
by acid cleavage and the product treated with dicyclohexylcarbodiimide in nitromethane to give 59 in 70% yield. Standard procedures were then used to convert this product to 5913, which proved to be identical with a sample prepared from ~ e p h a l o t h i n e . ~ ~ 3. THE TETRACYCLINES
The tetracyclines constitute an extremely useful and effective group of antibiotics. They are widely used in medical practice against a large variety of infective agents. The basic hydronaphthacene structure is shown in 60 and investigations into structure-activity relationships have revealed that considerable variation in R,, R,, R,, and R, can be made without much loss in its antibiotic activity. The naturally occurring derivatives of tetracycline (60a) are 7-chlorotetracycline (60b) and 5-hydroxytetracycline (60c). Two others are produced by a
3. The Tetracyclines
349
60a 60b 60c 60d 60e
60f
mutant strain of the original Streptomyes aureus. These are 6-demethyltetracycline (60d) and 7-chloro-6-demethyltetracycline (60e) Rl = C1; R2 = OH; R3 = R, = H). The latter two compounds are very resistant to both acid and base degradation, thus making them valuable for oral use. Other semisynthetic derivatives are also in common use. The most formidable of the synthetic problems posed by the tetracycline molecule have been discussed by W o o d ~ a r d . They ~ ~ . ~lie ~ principally in ring A : every carbon atom of this ring bears at least one substituent and three of the potentially six asymmetric centers of the molecule fall in the consecutive chain C-4, C-4a, and C-12a. Added to these difficulties apart from the three remaining asymmetric centers is the fact that the most highly substituted derivatives having both methyl and hydroxyl groups at C-6 are very sensitive to acid and base. These obstacles have not discouraged synthetic efforts and to date three groups have reported success. A.
6-Deoxy-6-Demethyltetracycline
The first success was reported by the Pfizer-Harvard group who in 1962 d e s ~ r i b e dthe ~ ~total ~ ~ ~synthesis of dl-6-deoxy-6-demethyltetracycline (60f). Their synthesis was accomplished in two phases. The first phase constituted the development of a reasonably efficient route to the intermediate 61, which
c1 I
61
was prechosen on the basis of its expected chemical versatility for phase two, the elaboration of ring A . The synthesis of 61, shown in Scheme 9, began with the base-catalyzed condensation of methyl 3-methoxybenzoate with dimethyl succinate or with
dd
GG
Q
350
3. The Tetracyclines
351
methyl acetate followed by alkylation of the intermediate @-ketoesterwith methyl bromoacetate. The product 62 was then condensed with methyl acrylate in a Michael reaction, using Triton-B as the catalyst, and afforded the keto-triester 63.The latter material was boiled with aqueous sulfuric acid to extrude the tertiary carbomethoxy group then reesterified to give the @-aroyladipate64.Hydrogenolysis of the keto group of 64 using a palladium catalyst gave by way of an intermediate lactone the half acid-half ester 65 (R, = H ; R, = CH,) which was purified by esterification and distillation of the resulting diester (65;R, = R, = CH,). Saponification then gave the pure diacid (65;R, = R, = H), which was chlorinated at the 6-position of the aromatic ring 66,thus ensuring that in the upcoming cyclization step, mediated by hydrofluoric acid, ring formation would give exclusively 67. The most crucial step in the sequence leading to 61 was the condensation of dimethyl oxalate with 67a togive 68.This was accomplished only after muchexperimentation and in view of the waywardness of the condensation must represent a kinetically controlled reaction. Success was achieved by using two equivalents of dimethyl oxalate almost four equivalents of sodium hydride and one equivalent of methanol in dimethyl formanide solution. Hot aqueous acid then smoothly transformed 68 into the much sought after triketone 61 and set the stage for the second phase (Scheme 10) of thekynthesis. Condensation of 61 with n-butyl glyoxylate in the presence of magnesium methoxide occurred smoothly to give 69. Treatment of 69 at -10" with dimethylamine afforded the Mannich base 70,which easily lost dimethylamine and consequently was reduced immediately after its preparation at -70" by sodium borohydride in dimethoxymethane. The product 71 was lactonized by means of a trace of acid in boiling toluene and the resulting compound 72 was reduced to 73 by means of zinc dust in formic acid (reaction time 1 minute) and then further to 74 by hydrogenation over a palladium catalyst. The final steps in the formation of ring A were accomplished by condensing the mixed isopropyl carbonic anhydride 75 with a new derivative of malonic acid, ethyl N-t-butylmalonamate (76).in the presence of magnesium ethoxide in acetonitrile. The product 77 was difficult to purify and in the crude state was treated briefly with sodium hydride in dimethylformamide containing methanol. This afforded the tetracyclic derivative 78 as a crystalline material which when treated with hot aqueous hydrogen bromide afforded the demethylated primary amide 79. The final step, the introduction of the 12a-hydroxyl group, was accomplished by means of carefully controlled oxygenation of 79 in the presence of cerous chloride in a buffered dimethylforrnamide-methanol solution. The product 60f was isolated only after an extensive purification procedure, but its identity was established beyond doubt by comparisons with the behavior of an optically active specimen derived from natural sources.
69
61
NMc,
CI / \
OMcO
53 % from 69
\
OH NMe,
70
90-95 %
wc@ 'OBu"
OH
OMcO
71
CI
{NMe,
C1
81 %
\
OMeO
\
%//
OH
/
0
c\
+
0 '
OMeO 73
NMc,
c I
H0
72%
4
OH
OMeO
OH
74
Scheme 10 352
\
\
72
NMe,
OH
I
OH
+
3. The Tetracyclines
353
NMe, /
CHaCONHBu'
C
I
\
\
OMeO
+I
OC0,CH Me2
COZEt
---+
OH 75
76
77
78
79
60f Scheme 10 (Continued)
An alternate synthesis of dl-6-deoxy-6-demethyltetracyclinehas been accomplished by Muxfeldt and Rogalski.28 They approached the problem by first constructing rings C and D and then simultaneously rings A and B. Their methods have the advantage of allowing the synthesis of tetracyclic compounds on a large scale, because of their relative simplicity. The synthesis (Scheme 11) has as its starting point 1-chloro-2-bromomethyl4methoxybenzene (80), which was used to alkylate the sodium salt of 1,1,2-tris(carbomethoxy)ethane. The resulting triester 81 was saponified to the corresponding acid which was decarboxylated at 160" to give the
+ $
CO,CH, C-CO2CH3
\
\
CH,
i
@Elr
OCHS
OCH3 CO,CH, 81
80
c1
CI
63
82
c!,r.p’ 84
85
H3C0,CTCONH-C(CH8)8
+-
CH30
0
0 87
86
86
Scheme 11 354
+-85 % (from 80)
3.
\
OH 0
355
89
H NH, H
WC'I
The Tetracyclines
---+
/
C-NH2
OH 0 90
Scheme 11 (Continued)
79
succinic acid 82. Polyphosphoric acid at 80" smoothly transformed 82 into 83. This was then converted2Bto the aldehyde 84 in excellent yield by a standard series of reactions which included protection of the ketone as the dioxolane, conversion of the carbomethoxy group by a chain extension sequence to a CH,CN group, then reduction of the latter to the aldehyde 84 with lithium aluminum hydride. Azlactone condensation of 84 with hippuric acid in the presence of lead acetate afforded 85, which when carefully hydrolyzed with hydrochloric acid resulted in the formation of 86, the key intermediate for the construction of rings A and B. This was accomplished in one step by condensing 86 with methyl N-t-butyl-3-oxoglutaramate(87) in the presence of two equivalents of sodium hydride in a mixture of ether and tetrahydrofuran. The product 88 proved to be a mixture of the two C-4 epimers, one of which could be isolated in a pure form. In order to remove the N-benzoyl group the mixture (88) was treated with triethyloxonium fluoborate and subsequently with aqueous acetic acid. This treatment yielded the pair of epimers 89. Demethylation and debutylation were accomplished simultaneously by means of hot hydrogen bromide in acetic acid. The resulting epimeric mixture, 90, was then methylated on the amine nitrogen by treatment with an excess of formaldehyde in methanol containing two equivalents of triethylamine and under the reducing conditions of a hydrogen atmosphere and a palladium catalyst. By this procedure, which also caused dehalogenation, 79 was obtained identical with the racemic material previously prepared by Conover et a1.26Oxidation of 79 to 6-deoxyd-demethyltetracycline (60f) was accomplished by means of oxygen and a freshly thus achieving the synthetic goal. reduced platinum
356
The Total Synthesis of Anlibiotics
B. Tetracycline
Tetracycline (60a) formally has been synthesized by Russian workers.31 In affect they synthesized racemic 12a-deoxy-5a,b-anhydrotetracycline 91,
91
which in its levorotatory form had been converted previously, in two steps, to tetracycline itself. Their synthesis commenced with a tricyclic intermediate 92 containing basically rings B, C , and D, which originally had been synthesized by Inhoffen et This was followed by modification of the functional groups of 92 to give the desired intermediate 93 on which was constructed the remaining elements of ring A. The synthesis3*of 92 is illustrated in Scheme 12. It was achieved in two steps from juglone 94, first by a Diels-Alder reaction with 1-acetoxybutadiene, which gave dominantly isomer 95, and then by reduction of the latter by one
94
95
92 Scheme 12
quarter equivalent of lithium aluminum hydride at -60'. Modifications3 of 92 (Scheme 13) began by preparation of its benzyl ether 96, which was then treated with excess methylmagnesium bromide
*..;; 3. The Tetracyclines
CHI
-
C8H5CH,0
/
0
__+
OH OH 98
357
C8H,CH20
OH 0 93
Scheme 13
to give 97, a base-catalyzed transfer of the acetate function also having taken place. Hydrolysis of 97 gave the trio1 98, which was easily converted to the desired dione 93. Condensation of the diendiolone 93 (Scheme 14)
99
100
101 R = 0 1Ola R = H lOlb R, = Phthal
I
PhCH,O Scheme 14
I
I1
OMc 0 102 R = Me 102a R, = Phthal
--*
103 R = Me 103a R, = Phthal
104 R 5 M e 104a R, = Phthal
H \
/
O
\
H
--3
CONH, PhCH,O 105
OMeO OH R = H, R' = COC,H,C02H-o
NMe, --*
91
-3
106
OH 0 Scheme 14 (Continued) 358
OH 0 60a
3. The Tetracyclines
359
with the triethylammonium salt of ethyl nitroacetate gave a mixture of two epimeric adducts 100. These when treated with dilute alcoholic hydrogen chloride underwent dehydration and afforded a single (?) nitro compound 101. Reduction of the latter by zinc in acetic acid then led to the amino ester 101a. At this stage the amino group was methylated with methyl iodide-silver oxide and the trimethylated product 102 hydrolyzed to the amino acid 103. However, under no circumstances could this acid, as the chloride, as the isopropyl carbonate, or as the isobutyl carbonate, be condensed with the magnesium salt of ethyl malonamate to give 104. In light of this the basic character of the amine function was for the time eliminated by the conversion of lola to the phthaloyl derivative lOlb by treatment with carboethoxyphthalimide. Methylation as before then gave 102a. When this compound was saponified by base, not only was the ester hydrolyzed, but the imide ring also was opened and had to be reclosed by heating in diglyme at 140". The resulting acid 103a was treated with phosphorus pentachloride in dimethylformamide to which was then added the ethoxymagnesium salt of ethyl malonamate. On this occasion the expected derivative 104a was successfully obtained. Cyclization of this compound by means of dimsyl sodium in dimethyl sulfoxide then led to the tetracyclic derivative 105, the imide ring again having opened. Hydrolysis by means of hot hydrogen bromide in acetic acid removed all of the protecting groups and when the resulting product was selectively methylated with methyl iodide in tetrahydrofuran the synthetic objective of the research 12u-deoxy-5~,6-anhydrotetracycline 91 was obtained. Except for the question of resolution of 91, this represents a formal synthesis of tetracycline because the conversion of 91 to 106 by oxygen and a platinum catalyst already had been reported3' by the Russians themselves and the final step a photooxidation had been accomplished by Schach von W i t t e n a ~using ~ ~ the procedure of Scott and Bedf~rd.~" C. Oxytetracycline
The synthesis of racemic oxytetracycline or terramycin (~OC),one of the most complicated and chemically sensitive of these antibiotics, was accomplished by Muxfeldt and his co-~orkers.~' The synthesis is based on the general methods developed previously for this type of tetracyclic system and which had culminated% in the total synthesis of 6-deoxy-6-demethyltetracycline (60; R, = R, = R, = R, = H). Oxytetracycline was assembled from three basic building blocks, the thiazolone 107, methyl 3-oxoglutaramate (108), and the aldehyde 109. The first two compounds are very easily prepared. The thiazolone 107 was
360
The Total Synthesis of Antibiotics
109
107
obtained as its hydrobromide by treatment of thiobenzoylglycine (110) with
110
107
phosphorus tribromide. Neutralization with sodium acetate then produced the free base, which proved to be very unstable and was best used immediately.38 The keto-amide 108 was prepared in two steps from dimethyl 3-0x0glutarate first by careful treatment with ammonia, which led to methyl 3-aminoglutaconamate, followed by acid hydrolysis of the latter compound. The synthesis (Scheme 15) of the aldehyde 109 had been achievedgDin part at an earlier date from the product 110 of condensation of juglone acetate with I-acetoxybutadiene. By contrast with juglone, which gives 95 (see above), juglone acetate gives as the major product of this Diels-Alder reaction the tetrahydroanthraquinone with the alcoholic oxygen at positions C-1 and C-5. Treatment of 110 with methylmagnesium bromide results in selective attack at the C-9 carbonyl group to give product 111 with the desired stereochemistry. This selectivity is undoubtedly due to participation of the C-1 acetate function as an ortho ester magnesium salt. Basic hydrolysis of 111 also caused isomerization at C-4a and afforded the trio1 112, which was converted to the ketal 113 by means of acetone in the presence of anhydrous copper sulfate. The cyclohexene ring was next degraded by glycolation of the double bond by means of a potassium chlorate/osmium tetroxide combination, followed by lead tetraacetate cleavage of the product to give the dialdehyde 114. Triethylammonium acetate then was used to catalyze the internal condensation of 114 to produce the unsaturated aldehyde 115. Ozonolysis of 115 followed by treatment of the crystalline ozonide with
__+
CHO AcO
0
CHO
118
Scheme 15
109
361
362
The Total Synthesis of Antibiotics
aqueous sodium carbonate yielded a mixture of two aldehydes 116 and 117, both of which could be isolated in a crystalline condition. That they are simply isomers about the carbon atom bearing the aldehyde group was shown by deuterium studies. In any event this mixture was treated with piperidike in boiling benzene and a single product, the enamine 118, was isolated, hydrolysis of the phenolic acetate function having occurred concurrently. The sodium salt of 118 was selectively alkylated with chloromethyl ether on the phenolic oxygen atom to give an enamine, which when adsorbed on deactivated silica gel underwent selective hydrolysis of the amine function and afforded the desired intermediate 109 as an oil. This hydrolysis was not only selective but also stereospecific since 109 had an NMR spectrum consistent only with the structure depicted and in addition when treated with acetic acid regenerated 117 in almost quantitative yield. The synthesis of oxytetracycline was now completed as shown in Scheme 16. The condensation of 109 with 107 occurred without any isomerization
WHO Ys CHI
CH3
oxo
H
CH 3 ‘Ill,
N=fCH,
+
0
0
CHjOCH20
qy
107
109
N+ceH,
Cli03 ~ : t ’ 3
CH30CH,0
71 %
__t
+ CH30,CCH,COCH2CONH2 21%_
t+s 0
0
119
CH3,/ 0-0
108
CH,
N t4CSC6H, 32 %
__t
CONH, 120 12On
R = CHSOCHZR=H Scheme 16
3. The Tetraclines
363
60c
Scheme 16 (Continued)
when basic lead acetate was used as the catalyst in tetrahydrofuran. In a masterful stroke the product 119 of this condensation was doubly condensed'O with methyl 3-oxoglutaramate (108) using a combination of butyllithium and potassium I-butoxide as the catalyst to give the tetracyclic product 120. Introduction of the 12a-hydroxyl group and the dimethylamino function were now carried out sequentially. The former was accomplished by oxygenation of 120a (obtained from the parent methoxymethyl ether 120 by acetic acid catalyzed hydrolysis) in a basic medium, giving, after acid hydrolysis, the thioamide 121. The latter compound was methylated on sulfur by treatment with methyl iodide at room temperature in tetrahydrofuran and the intermediate thioimino ether iodide hydrolyzed in acid without isolation to give N,N-bisdemethylterramycin(122) as the hydrochloride. The latter was immediately alkylated with a combination of methyl iodide and Hiinig's base in tetrahydrofuran. Purification of the product by chromatography on a polyamide substrate then afforded dl-terramycin to mark the end of a brilliant and highly successful 10-year program directed to the synthesis of this compound. Other groups have attempted total synthesis in this area. Most notable of these efforts are those due to Kende et which culminated in the synthesis
364
The Total Synthesis of Antibiotics
of 123, and the more recent work of Barton and his collaborators who have
123
124
published42only very brief details of the synthesis of 124. The latter differs from previous syntheses in that both rings A and D are aromatic throughout the elaboration of the tetracyclic nucleus. A reaction now needs to be found which will in effect introduce the 12a-hydroxyl group and eliminate the aromatic character of ring A. 4.
THE BASIC SUGARS
Several basic sugars produced by differing species of Streptomyces have useful antibiotic properties. Some have similar activities while differing in toxicity, but generally they are most effective against gram-negative bacteria and tubercle bacilli in addition to gram-positive bacteria. Hepatotoxicity and ear damage are the chief drawbacks to their use.
N-Methylglucosarnine
Streptose
Streptidine
H 125
4.
The Basic Sugars
365
In structure there are some similarities but the one thing they all have in common is that they contain unusual amino sugars. The oldest of them in use is streptomycin-A (125) isolated4a first in 1947. Despite this, the total synthesis of this molecule has not been achieved, although the syntheses of the individual components ~ t r e p t i d i n e , ~N-methylglucosamine,4" ~ and streptose4'Ihave been reported. More closely related are gentamicin-A 126, the kanamycins A (127a), B (127b), and C (127c), and the more complicated paromomycin I1 (129a) and neomycin B (129b). All of these antibiotics contain the cyclohexane derivative 2-deoxystreptamine. In addition the amino sugar glycosides of 2-deoxystreptamine, paromamine (128a) and neamine (128b), are contained,
Paromamine (128% R, = OH; R, = NHJ Neamine ( l a b ; R, = RE = NH2) 2-Dcoxyst rcptaminc
0
I
NH,
Kanamycin A (127a; R, = NH,; R, = OH) Kanamycin B (127b;R, = R, = NH,) Kanamycin C (127c; R1 = OH; R2 = NHz)
366
The Total Synthesis of Antibiotics
Paromomycin (129~;R = OH) Neomycin (129b; R = N H b
respectively, in paromomycin I1 and kanamycin C, and neomycin B and kanamycin B. On the other hand, gentamicin A (126), which is the major component of a complex of at least 15 compounds, differsq7 only from ring. In this ring, in gentakanamycin C in the 3-amino-3-deoxy-~-glucose micin A the 3-amino group is monomethylated and the hydroxymethyl group has been replaced by hydrogen. It is in fact a D-xylo-pyranose ring. More distantly related to these compounds is spectinomycin4* (130), which contains a streptamine ring, similar to that of streptomycin, but without the guanidino functions.
130
Still more remote in structure is kasugamycin (131), which sees use mainly against rice blast and has no useful therapeutic effect in mammalian species. Completely unrelated to any of the foregoing compounds are lincomycin (132a) and streptozotocin (133). The former compound has potent antibacterial properties but its semisynthetic derivative (132b) is more useful
4. The Basic Sugars
367
therapeutically, especially in the treatment of malaria. Streptozotocin, a glucosamine derivative, is used experimentally as an antibacterial and antitumor agent.
HO
CH,OH
cH8b H% :oH
NHCONCH3
I
HO
132a R, = OH; R, = H 132b R, = H; R, = CI
OH
133
NO
Of these antibiotics only the kanamycins, kasugamycin and streptozotocin have been synthesized. The extent to which protection of functionality must be taken in the synthesis of amino sugars is probably greater than that with any other group with the possible exception of the peptides. This represents both an experimental and psychological barrier to total synthesis in this area. A. The Kanamycins The key compound in the synthesis of the kanamycins is 2-deoxystreptamine (134). Its synthesisqewas first reported in 1964 and is illustrated in Scheme 17. Hydrolysisso of the epoxy-diacetate 135 followed by acetylation afforded the tetraacetate 136, which after subsequent treatment with bromine-water led to the bromohydrin 137 (via axial addition). Debromination of 137 with Raney
368
The Total Synthesis of Antibiotics
OAc
B I'
OAc 65 %
+
c_f
=_ -
OAc 135
OAc
'-lo
dAC
140
OAc
OAc 137
136
138
82 %
__j
139
141
nickel followed by ammonolysis to remove the acetyl groups gave 3-deoxyepiinositol (138). Catalytic oxidation of 138 using oxygen and a platinum catalyst, specific for the oxidation of axial alcohols, yielded the monoketone 139 whose oxime 140 was reduced by means of sodium amalgam to 141 isolated as its pentaacetyl derivative. Ammonolysis of the latter afforded 142, which was subjected to the same sequence of reactions used with 138. This resulted in the pentaacetyl derivative of 134. Racemic 2-deoxystreptamine (134) itself was liberated from this derivative by hydrolysis with 4N HCI and the subsequent use of an anion exchange column to obtain the free base. Kanamycin A (127a) has been synthesized independently by NakajimaK1 and by Umezawa.6a The former workers prepared a protected derivative of the 6-amino-6-deoxy-~-glucose moiety, coupled it with a masked derivative of
4.
The Basic Sugars
369
2-deoxystreptamine, and then forged the second glycosidic linkage by means part of the molecule. of a protected form of the 3-amino-3-deoxy-~-glucose Umezawa and his associates did just the reverse, but both groups needed roughly the same protected components. The key intermediates chosen by Umezawas3 for the first stage in the synthesis were 3-acetamido-2,4,6-tri-o-benzyl-3-deoxy-a-~-glucopyranosyl chloride (143) and the isopropylidene derivative of N,N’-bis(carbobenzy1oxy)2-deoxystreptamine (144). The former compound was prepared (Scheme 18) by a method resting on some earlier work by B a e P in which methyl b-D-glucoside was oxidized to the dialdehyde 146 by means of periodate. Base-catalyzed condensation of 146 with nitromethane then produced the nitrosugar 147, which when reduced catalytically gave the desired amino sugar 148. Acetylation of 148 with acetic anhydride followed by benzylation with benzyl bromide in dimethylformamide in the presence of both barium oxide and hydroxide led to 149 in high overall yield. Hydrolysis of 149 with sulfuric acid followed by reacetylation afforded a mixture of anomers 150, each of which gave the sought-after intermediate 143 in quantitative yield when treated with dry hydrogen chloride in dioxane containing acetyl chloride. The second component, 144, in its racemic form, was easily prepared by the action of 2,2-dimethoxypropane in the presence of an acid catalyst on
N,N’-bis(carbobenzyloxy)-2-deoxystreptamine.
Condensations6 of 143 with 144 in the presence of mercuric cyanide and Drierite in anhydrous dioxane/benzene gave a crude product in 80% yield which was hydrolyzed with 80% acetic acid to split the acetonide and hydrogenated over palladium to remove the benzyl groups. The residual material was dinitrophenylated and the product chromatographed to give two components, of which one was the desired diastereoisomer 151. i t was identified by comparison with a specimen prepared from a degradation product of kanamycin A. Hydrolysis of 151 with methanolic ammonia then afforded the free aminosugar 152, signaling completion of the first phase of the synthesis. In the second phase the amino groups of 152 were protected by carbobenzyloxylation and the derivative, 153, was then converted to its bisacetonide. The latter when benzylated afforded 154. Acetic acid cleavage of the ketal rings followed by treatment of the resulting tetrahydroxy derivative with 2,2-dimethoxypropane at 5” successfully yielded the monoacetonide 155. Condensations2 of 155 with 2,3,4-tri-o-benzyl-6-(N-benzylacetamido)6-deoxy-a-~-glucopyranosyl chloride (156)53*5sunder Konigs-Knorr conditions, as in the formation of the first glycosidic linkage gave a brown viscous mixture. This was treated with 80 % acetic acid, then hydrogenated over palladium to remove U-benzyl groups, de-N-acetylated with barium
!5i
0
370
CH,OC,H
NIICO,C,H, 12%
f
__+
NHCO,C,H,
HO 143
144
NHR
84 % _ j
151 R = 2,4-NOZC,H, 152 R = H 153 R = CO,C,H,
NHCOZCYH,
N HCOZ C7 H;
n
+ CH3 155
Scheme 19
156 371
372
The Total Synthesis of Antibiotics
NHRl
CH20R2
R20+ RlNH 157
127a
OR2
R, = 2, 4-(NOz)zC6H,: Rz = AC R, = R, = H Scheme 19 (Continued)
hydroxide solution, then hydrogenated again to remove the N-benzyl group. The product of these transformations was dinitrophenylated then o-acetylated The residual material, a six-component mixture, was chromatographed on a silica gel column and the main product 157 was isolated in 10.1% yield overall from 155. When the latter material was treated with methanolic ammonia and the free base isolated by chromatography over an ion-exchange resin, crystalline kanamycin A was obtained, in all respects identical with the natural material. In the alternative synthesis (Scheme 20) of 127a due to Nakajima,61 6-acetamido-2,3,4-tri-O-benzyl-6-deoxy-a-~-glucosylchloride (158) was condensed with racemic 144 in a modified Konigs-Knorr reaction and after subsequent removal of the isopropylidene group afforded a mixture of the two diastereoismers 159 and 160 in 34 and 40% yield, respectively. A second condensation under essentially the same conditions this time between 159 and 143 produced the highly protected derivatives 161 and 162. Removal of the benzyl and carbobenzyloxy groups from 161 was accomplished by sodium in liquid ammonia at -70". This was followed by N-acetylation and led to tetra-N-acetylkanamycin A, completely identical with a specimen obtained from the natural product. Hydrolysis of the tetraacetyl derivative by barium hydroxide then afforded kanamycin A (127a) itself. When 162 was subjected to the same treatment it also gave a tetraacetyl derivative isomeric with that derived from kanamycin A but whose NMR spectrum suggested that the new glycosidic linkage had the a-configuration,
+
373
YP
1
I
+ 143
Kanamycin A 127a
161
159
Scheme 20 (Continued)
162
4. The Basic Sugars
375
and that it had formed at the C-5 oxygen of the 2-deoxystreptamine moiety. The kanamycins B (127b)and C (127c)were both synthesized by Umezawa and his co-workers in a coupling sequence opposite to that used in their synthesis of kanamycin A. In addition both syntheses depend on the initial synthesis of paromamine (128a),which was used for the preparation of both kanamycin C (127c) and neamine (128b). The latter was then used to synthesize kanamycin B (127b). The synthesis5’ (Scheme 21) of paromamine (12811)was carried out by
CH,OAc
DNPHN 164
Sr
0H
-1
163
1668 R1= OH; R, t= AC 166b R, = OTS; R, = Ac 166c R, = NHAc; R, = AC l2Sb R, = NH,; R, = H
Scheme 21
NHDNP
376
The Total Synthesis of Antibiotics
condensing bis-N,N'-(2,4-dinitrophenyl)-2-deoxystreptamine (163) with the protected a-glucopyranosyl bromide (164) derived68 from glucosamine. The bromide was used here because of the comparative lack of reactivity of 163, and the condensation was carried out in nitromethane at 95" for 15 hours in the presence of mercuric cyanide and bromide. This gave a mixture of glycosides which were acetylated and separated by chromatography. The correct diastereoisomer, 165, was obtained in low yield. Hydrolysis of 163 with methanolic ammonia followed by treatment with an exchange resin gave crystalline paromamine (128a). The conversion of paromamine (128s) to neamine (128b) proceeded via the tri-N-acetyl derivative 166a, which was selectively monotosylated in 43 % yield to give 166b. The tosyloxy group of this compound was then successively converted to the azide, the amine, and the acetylamino compound 166c by standard procedures. Deacetylation of 166c by means of hydrazine then afforded, after the usual purification procedures, neamine (128b) in 72% yield. ) ~ ~ then completed The synthesis of kanamycin B (127b)59and C ( 1 2 7 ~ were by coupling, respectively, the protected derivatives 167 and 168 of neamine and paromamine with the a-D-glucopyranosyl chloride 143. The syntheses parallel the methods used for the total synthesis of kanamycin A and will therefore not be discussed in further detail here.
167
4.
The Basic Sugars
377
B. Kasugarnycin The total synthesis of kasugamycin (131) has been accomplished by H. Umezawaal (not to be confused with S. Umezawa, whose group synthesized the kanamycins), who synthesized the aminosugar moiety beginning with a pyranone derivative. An alternate synthesis due to Nakajima reporteda2 slightly earlier, started with a D-glucose derivative and afforded the naturally occurring optical isomer directly. 71 % __f
0 169
170
171
70 % __f
tH3
173
172
175
174
176 Scheme 22
In the synthesis (Scheme 22) due to Umezawa and his group the starting dihydropyranone 169 was treated with nitrosyl chloride at -60" and afforded the dimer of 170. Hydrolysis of 170 by water led easily to 4-oximino-Soxohexanoic acid (171), which then was hydrogenated stereoselectively over a platinum catalyst to d/-erythro-4-amino-5-hydroxyhexanoic acid (172). Lactonization of 172 at room temperature with acetic anhydride afforded 173, which was reduced to the hemiacetal 174 by lithium aluminum hydride. Refluxing with acetic anhydride converted 174 to 175, and chloronitrosation of the latter compound under conditions similar to those used for 169, afforded the dimer of 176. Coupling of 176 with excess 1:2, 3:4-di-oisopropylidene-(+)-inositol (177) was achieved in methylene chloride at 0" in the presence of silver carbonate and perchlorate and led to a mixture of diastereoisomers which were reduced over a platinum catalyst. The isopropylidene groups were removed by boiling 50% acetic acid and the
378
The Total Synthesis of Antibiotics
+ 176
(IO.>%crudc) (1 % pure)
177
R = Ac 178b R = H 131 R = C(NH)CO,H 178a
Scheme 22 (continued)
OH
reaction was product purified by chromatography over an ion-exchange resin and by crystallization to give 178a. Hydrolysis of the acetylamino group by barium hydroxide then afforded optically active kasuganobiosamine (178b).The synthesis of 178a is noteworthy not only from the point of view of the formation of the a-glycoside but also because of the resolution that took place during the isolation procedure. The completion of the synthesis of 178b in effect completed the synthesis of 131 because the final step, involving the reaction of 178b with diethyl oxalimidate followed by acid hydrolysis under mild conditions, had been reporteda3 previously. Nakajimaal in his synthesis (Scheme 23) of 178b started with the 3-deoxyglucopyranoside 179a easily prepareda4 from D-glucose in five steps. Conversion of 179a to 179b was achieved in three steps involving a displacement with inversion of the tosyl group by azide ion followed by catalytic reduction over a platinum catalyst then acetylation. The benzylidene group of 179b was removed by means of warm aqueous sulfuric acid and the resulting oil was monotosylated then acetylated to give 180a. Replacement of the tosyl group by iodide and reduction with Raney nickel led to 180b, which was deacetylated by sodium methoxide in methanol. The product 181 had its 4-hydroxyl group in the wrong configuration for further work and this was reversed by chromic acid oxidation to the 4-0x0-compound followed by platinum-catalyzed hydrogenation to give 182a. The latter compound was converted by mesyl chloride to 182b, which was then transformed into 183a by the same procedures used for the conversion of 179a to 179b. Hydrolysis with 5N formic acid followed by acetylation led to 183b, which was treated with a saturated solution of hydrogen chloride and acetic acid/chloroforrn to produce 184. Coupling of 184 with 177 gave a crystalline glycoside 185.
4. The Basic Sugars
COH,
-
CH,R
$ - +-oo
___, 55g: R2
qAc
AC0’;”t
CH3
OCH3
179a R, = H; R, = OTs 179b R, = NHAc; R, = H
OCH, 181
R = OTs R =H
ACHN9C
(3-43
--c
184
lROa 180b
RO
c
H 0 - q
379
0 -
d
OCH3
l82a R = H 182b R = MeSO,
CH3
183a 183b
OR R = CH, R = AC
-cI
+178b
185
Scheme 23
Hydrolysis of 185 followed by acetylation provided a heptaacetyl compound, which when deacylated first by sodium methoxide to remove 0-acetate groups then by boiling aqueous barium hydroxide to cleave the amide groups, led to kasuganobiosamine (178b), thus constituting, in view of previous work,Eaa total synthesis of 131.
C. Streptozotocin This antibiotic was first synthesizede6at the time its structure was deduced. This was accomplished by treating tetra-a-acetylglucosamine hydrochloride
380
The Total Synthesis of Antibiotics
(186a) with methyl isocyanate to give the unsymmetrical urea 186b. Nitros-
186a R = NH,HCI 186b R = NHCONHCH, 1 8 6 R = NHCON(NO)CH,
dON(NO)CH, 133
ation of 186b in pyridine produced 186c, which was deacetylated by means of ammonia in methanol at - 10". This afforded streptozotocin (133)identical with an authentic sample. A preparative method for the synthesis of 133 has also been published.Ge This involves a direct reaction of D-glucosamine with N-nitrosomethylcarbamylazide and gives streptozotocin in 31 % yield. The nitrosoazide CH,N(NO)CON, was prepared in good overall yield from niethylisocyanate by the following sequence: CH,NCO
--t
CH,NHCOCI
--+
CH,NHCOCl
+
CH3N(NO)CON3
D. Lincomycin This antibiotic was first synthesizedGGa in a partial manner from its cleavage (187) and the amino sugar products trans-1-methyl-4-n-propyl-L-proline methyl a-thiolincosaminide (187a). The latter had been isolated during previous degradation studies.6sb
187
These reactions are outlined below and utilize as the first step a variation of a procedure that Kenneret al.sechad used for the synthesis of I-carbobenzoxy4-methylene-L-proline. Hydrogenation of the intermediate propylidene
4.
The Basic Sugars
381
derivative 187b over a platinum catalyst on Dowex resin then gave a mixture of the desired N-blocked 4-n-propylprolines (187c) containing 25-35 % of the desired trans-isomer. Catalytic reduction on other supports afforded little or none of this isomer. Coupling of 187c with 187a was accomplished using the mixed anhydride method (isobutyl chloroformate triethylamine in dry acetonitrile) and led to 187d in high yield. The blocking group was removed by hydrogenolysis, and the resulting product was reductively methylated using formaldehyde and a palladium-on-carbon catalyst in the presence of hydrogen, This gave a mixture of lincomycin 132a and cislincomycin (187e) which were separated by careful chromatography.
+
Cbz
I
Cbz
+ Ph3PCHCHZCH3
55 %
I
__+
A 0 2 H CZHBCH 187b
(Standard sugar nomenclature is used For the sugar side chain in these and subsequent diagrams of this section.)
The Total Synthesis of Antibiotics
382
Alternatively the mixture of these compounds could be prepared by deblocking 187c, reductively methylating the nitrogen atom, followed by coupling with 187a via the mixed anhydride method. The total synthesis of 132a then devolved simply on a synthesis of the amino sugar 187% This was accomplished both by Magerleinsed and by Szarek et al., the latter workers devising two different syntheses of the molecule. Contrary to the methods evolved by the Canadian group Magerlein introduced the S-methyl group at the outset. His synthesis of 187a is outlined below. Treatment of D-galactose with methanethiol in hydrochloric acid afforded 187f, which by a controlled tosylation reaction was converted to 187g (R = H). Acetylation of the latter compound gave an oily triacetate 187g (R = Ac), which underwent substitution with sodium iodide to give 187h. Replacement of the 6-iodo by nitro was slow (sodium nitrite in dimethylformamide) but afforded about 20 % of the desired compound 187i. Repeated additions of acetaldehyde and sodium methoxide to a methanolic solution of 187i then led to the nitro-alcohols 187j in 50% yield. The latter
HO
OAc
HO
--+
1871
SCH
,
187g
AcO
SCH,
5
187h
SCH,
CH3
I
HO THOH
AcO A
c
O
q
187i
2
--t
SCl-13
HO
187j
-+ (187a)
SCH 3
when reduced with lithium aluminum hydride in tetrahydrofuran afforded a mixture of amino sugars containing methyl a-thiolincosaminide. The desired isomer was separated as its pentaacetyl derivative by chromatography over silica gel.
4. The Basic Sugars
383
In the alternative syntheses due to Szarek and his associatesaae the thiomethyl group was introduced towards the end of the synthesis. The first method involved the sequence shown below, in which initially 1,2: 3,4-diO-isopropylidene-a-D-galactohexodialdopyranose-(1,5) was condensedaa* with ethylidenetriphenylphosphorane to give predominantly the cis-isomer (187k) of the desired oct-6-enose. Treatment of the latter with aqueous CH3
CH3
CH3
%$--*
C H f q H V c H 3 1P?L.
m-'m
%O
CH3 CH, CH3
BzO- r H
Ro+H
18"
%0 C H 3 CH3
P'
187111
/O C H 3PqU -
%O
CH3 CH3
I
I,o
CH (T CH3
1870
18'ln
0
4
384
The Total Synthesis of Antibiotics
potassium permanganate gave the crystalline diol 1871 (R = H)from which the monobenzoate (1871;R = COPh) was easily prepared. Oxidation by the Pfitzner-Moffatt reagent then afforded the ketone 187111 (X = 0), which was converted to a mixture of the oximes (187111; X = NOH) in which the anti-form predominated. Reduction ofthe latter isomer with lithium aluminum hydride led to a mixture of the erythro and threo forms of 187n. Separation of the former in low yield was achieved with difficulty by means of chromatography. An alternate route to 187n involved debenzoylation of the oxime 187m (X = NOH) with a catalytic quantity of sodium methoxide in methanol followed by reduction of the oxime function by Raney nickel. Acetylation then afforded 18711,but this constituted mainly the undesired threo isomer.
H3 c --+
-
187q
H 3C
0 NHAc
--+ (18714
CI1,
4. The Basic Sugars
385
Treatment of erylhro-187n with methanethiol and concentrated hydrochloric acid afforded the dimethyl dithioacetal 1870 which treated with dilute acid yielded the N-acetyl derivative of thiolincosaminide. Deacetylation was achieved in high yield in refluxing 9 5 % hydrazine hydrate and afforded 187a identical with the material derived from lincomycin. The key intermediate in the above synthesis, namely the erythro form of 18711, was also synthesized using the sequence described. The condensation of the starting aldehyde with nitroethane in the presence of sodium methoxide gave a mixture of 8-nitro-alcohols, which were acetylated to give the corresponding 8-nitro-acetates 187p of which three were obtained pure by fractional crystallization. The predominant isomer when refluxed with triethylamine in benzene afforded the cis- and /runs-isomers of 187q in a ratio of 6: 1. When the cis-isomer was treated with saturated methanolic ammonia, a mixture of the two stereoisomeric uic-nitroamines (187r; R = H) resulted. N-Acetylation of this mixture produced the corresponding acetyl derivatives (187r; R = Ac) in a 1 : 1 ratio. Oxidative denitration of this mixture with potassium permanganate led to two acetylamino-ketones the minor one of which (187s), when reduced with sodium borohydride gave two uic-acetamido alcohols. Fractional crystallization allowed their separation, and one proved to be identical with the previously prepared 187n. An alternate but somewhat lengthier procedure for the preparation of the intermediate N-acetyllincosamine, the hydrolysis product of 187n, also has been published recently by Japanese workers."" Again, the starting point in their synthesis was the 1,2,3,4-di-O-isopropylidene-a-~-gu~uc~o-hexodialdo1,5-pyranose used by Sarek and his associates.'8C The Japanese workers treated the fatter with sodium cyanide in aqueous methanol and obtained a mixture of the L- and D-glycero cyanohydrins 187t, which proved to be difficult to separate. Tosylation then gave the O-tosyl compounds, which were easily separated, and treatment'" of the D-glycero isomer 187u with lithium aluminum hydride followed by acetylation afforded the aziridine 187v. In contact with warm acetic acid 187v afforded the 6-acetamido-7-0acetate [187w (R - Ac)] exclusively. Deacetylation by means of sodium methoxide in methanol led to the 7-alcohol derivative 187w (R = H). Oxidation of the latter by the Pfitzner-Moffatt procedure generated the aldehyde 187x (R = Ts), which when treated with methyl magnesium bromide produced the undesired threo isomer 187y. Conversion of this material to the desired isomer 1872 was achieved by oxidation of 7-hydroxyl group to the ketone by means of chromium trioxide in pyridine followed by sodium borohydride reduction. Hydrolysis of 1872 with aqueous acetic acid or the acid form of Amberlite IR-120 then afforded the desired N-acetyllincosaniine.
Ac
Ji"YHO AcNH
AcYH
H3C A-H _+
H3C
__*
CHB AcNH\
-H
I I
C--M
C-H
H __*
OH 1872
386
N-acetyiiincosamine
5. Nucleoside Antibiotics
387
5. NUCLEOSIDE ANTIBIOTICS
About half of the therapeutically useful members of this group are closely related purine nucleosides. Angustmycin A (189) is the dehydration product of psicofuranine (188), both of which are derivatives of D-psicose
HOCH2 R
HO N HOH2 0 H 188
c H 2 HO $ ~ OH z~~
HO OH
189
190
Q Q
HOCH, K
OH
tiOCH2 R'
H N OH
191
NH, 192
OH
R'=
yt;>I
Nucleocidina7(190), on the other hand, is a ribofuranose derivative with the most unusual fluoro- and sulfamoyloxy groups on the sugar nucleus. It is related to cordycepin (191), an antibiotic of little medical value first isolateds* from the culture fluids of Cordyceps milituris (Linn) link. In this group also is puromycin (192), a 3-amino-3-deoxyribofuranosyl purine. The most
388
The Total Synthesis of Anlibiotics
unusual member of the group is septacidineg (193), in which the 4-amino-4deoxy-L-glycero-L-glucoheptoseis attached at the 6-amino group of the adenine nucleus. Three other compounds discussed in this section, tubercidin (193a), sangivamycin (194), and toyocamycin (195), are all closely related pyrrolopyrimidine nucleosides. These substances and the purine nucleosides described above all have been synthesized with the exception of nucleocidin and septacidin.
193a 194 195
R=H R = CONH, R = CN
Hi) O H
The remaining compounds, excepting the racemomycins, are peptide nucleosides derived from pyrimidine. The p o l y o ~ i n s which , ~ ~ comprise an antibiotic complex of 12 compounds, are nucleosides derived from uracil. Polyoxin A, for instance, has structure 196. These compounds have great
H,N-CH
I I HO-CH I
CII-OH
CH,OCONH, 196
commercial importance in the Far East for the control of the sheath blight disease of the rice plant. None of the polyoxins has been synthesized but
5. Nucleoslde Antibiotics
389
synthetic work on the sugar residue has been reported.'' Blasticidin S (197a) is a cytosine nucleoside and is obviously related to gougerotin (197b), another nucleoside antibiotic,'l" and to amicetin (198),72an antimicrobial
CH3
NH, 197a
CH20H I
197b
compound having little or no therapeutic use at this time. Racemomycin A
198
'CH20 H
is extremely similar if not identical to ~ t r e p t o l i n(199), ~ ~ perhaps the only question being the position of the carbamate group in the former. The structure of racemomycin 0 is again a variation of structure 199, but the position of an additional CH,CH(OH)CH,OCH,CH,CH< fragment has
not been ~larified.'~ No total synthesis work has been reported on these compounds. In general the syntheses reported here have followed the well-established techniques for the synthesis of nucleosides wherein almost all of the work
390
The Total Synthesis of Antibiotics
I
H,N(CH,),CHCH,CO
I
NH, 199
involves the modification of a readily available sugar (or derivative) via blocking-deblocking methods, to arrive at the required sugar moiety. A.
(I).
Purine Nucleosides Puromycin
Puromycin was the first of these antibiotics to be ~ynthesized~~ and demanded that a suitably protected derivative (214) of 3-amino-3-deoxy-~-ribofuranose be prepared prior to the attempted formation of the nucleoside bond. This was a c ~ o m p l i s h e das~ ~shown in Schemes 24 and 25, starting with o-xylose (200), which was initially converted to a mixture of the methyl glycosides (201). Under very carefully controlled acid conditions (2 x N sulfuric CH,OH
CH2OH
QOH
87%,
Q-
OCH,
OII
'.$-
OH 201
200
CH3
OCH
53-57%
B 100%
~ 9 6 %
CH3 H3& (
OCH3 OMS
OH
203
202 Scheme 24
___,
5. Nucleoslde Anllblotiw
-
CHpOH
@OCH3 + ;;;z C&
OCH
0M5
205
204
p
391
CHpOH CH,OH
CH,OfI
0:
W O C H , 4- b O ( 3 1 3
NH,
NI-I, 206
54%
OCH3
5 NII
ago%
B 100% d
/"\
H3C
207
CH3
pocH3% @ 208
CH,OH
AcOCH,
MsOCH~
OCH,
NHAc 209
AcNH 210
94%
V
O
C
H
OAc
AcNH
211
Scheme 24 (Continued)
acid) in an excess of acetone this mixture was transformed into the monoisopropylidene anomers 202, which could be separated fairly easily (33 and 21 % yield) because of a large difference in their boiling points. The remainder of the synthesis was carried out on each anomer individually, in an identical fashion. Each series eventually afforded an anomer (214) that could be used in the final steps of the synthesis. Treatment of 202 with mesyl chloride afforded 203, which was split to the diol 204 by means of warm 70% acetic acid. Sodium methoxide then was used to generate the anhydroglycoside 205. The latter compound reacted with ammonia to give a mixture of the aminoglycosides 206 and 207, but in the case of each anomeric series the addition of acetone caused the separation in crystalline form of the desired imine 208. Aqueous acetic anhydride converted 208 to 209, which had the 2-hydroxy group in the wrong orientation. This was reversed in two steps, first by conversion to the dimesylate 210 and then by treatment of 210 with sodium acetate in boiling 95 % methyl cellosolve. The product of the reaction was in
3
The Total Synthesis of Antibiotics
392
fact not 211 but the corresponding noncrystalline 2-hydroxy compound, which was converted to the highly crystalline 211 by means of acetic anhydride pyridine. The intermediate alcohol arises because of neighboring group participation of the acetylamino group which gives an oxazoline ring that subsequently is hydrolyzed by the water present in the reaction mixture. In the next phase of the synthesis (Scheme 25) the triacetyl derivative was 0-deacetylated by sodium methoxide in methanol to 212, which under normal benzoylating conditions afforded the dibenzoate 213. Either anomer of 213 when treated with hydrogen chloride in acetic acid under well-defined conditions led to the desired protected sugar 214. CH,OAc @-.0CH3
CH,OH
loot.._
b o c I i 3% b
1
1 OAc NHAc
CH,OBz C
H
3 __t % ’
I ouz
OH
N H Ac
NHAc
211
O
212
213
FH,OBz
CH,OBz I
~ N H Ac
O
A
C
NHAc
NHA~
2 14
215
218
216
219 Scheme 25
2 17
Antibiotics
80 % +
393
HocP 64 %
__t
NH,
NHAc 220
N(CH 3 1,
__+
NH
NH
I co
I
co
'
H-C-CH2
I
NHR
'
H-C-CH,
a 222
O
C
H
R = OCOC,H,
3
I
--(J-0CH3
NH, 192
Scheme 25 (Continued)
Attempts to prepare the chlorosugar 216, needed for the coupling reaction with the purine derivative 217, were not initially successful. However, it was found finally that titanium tetrachloride would react with 215 to give 216, with which it also formed a complex. The reaction of this complex with chloromercury 2-methylmercapto-6-dimethylaminopurine(217) in boiling ethylene dichloride for 18 hours afforded the desired protected nucleoside 218 in high yield. Raney nickel desulfurization then eliminated the methythio group to give 219. Removal of the sugar protective groups was achieved" by treatment first with sodium methoxide/methanol to give 220 then with aqueous barium hydroxide to effect N-deacetylation. The product was 221, the necessary nucleosidic portion of puromycin. Introduction of the amino acid was then accomplished by the reaction of 221 with the carbethoxy mixed
394
The Total Synthesis of Antibiotics
anhydride of N-carbobenzyloxy-p-methoxy-t-phenylalanine,which led to N-carbobenzyloxypuromycin (222). Hydrogenolysis of 222 with a palladiumon-charcoal catalyst in acetic acid then afforded puromycin (192), identical in all respects with the natural product. (2).
Psicofuranine and Angustmycin A
Psicofuranine 188 is a nucleoside comprised of adenine and D-psicose. It was first synthesized, essentially from these components, by the group of investigators that deduced its str~cture.'~ The D-psicosyi chloride tetraacetate 226 needed was prepared according to the procedure of Wolfrom et a1.,7* (Scheme 26) in which ribonyl chloride tetraacetate was treated with diazoH(CH,OAc),COCI 223
-%H(CH,OAc),COCHN,
so %
+
224
H(CH,OAC)~COCH,OAC91%_ 225
AcOCH,
AcO OAc
HgCl 227
226
NH,
NHAc
I
I
228
Scheme 26
188
methane to give 224. Reaction of 224 with acetic acid gave keto-D-psicose pentaacetate (225), which with hydrogen chloride in ether afforded the
5. Nucleoside Antibiotics
395
required chloride 226. Condensation of 226 with chloromercuri-6-acetylaminopurine (227) provided 228 in unspecified yield. Deacetylation then afforded psicofuranine (188) identical with the natural product. A second synthesis of 188 has also been recorded by Farkag and Sbrm.80 In their case D-psicosyl bromide tetrabenzoate was condensed in dimethylacetamide with the chloromercury salt of 6-benzoylaminopurine and the product then debenzoylated by means of barium methylate solution. After chromatography of the crude material psicofuranine was obtained in 4.6 % yield. Angustmycin A (189) has been synthesizeds1partially from psicofuranine. In this transformation it was necessary to protect the 1,3- and 4-hydroxyl groups while a method was found for the dehydration of the 6-hydroxyl. This was accomplished as shown in Scheme 27.
pd-
IOCHp
CH20H
110 011
'18 %
65%
O&
H OC2H5
188
-
TsOH9C
C
O& 231
Qi9-
HOCHp
230
229
H
p
G
O&O
dd JO+ 62% ( ==pHC
CH2
232
CH2OH HO OH 189
Ad = Adenine
Scheme 27
Treatment of psicofuranine (188) with triethyl orthoformate at room temperature afforded the ethoxymethylidine derivative 229, which underwent a facile ring closure to the orthoester 230 when subjected to boron trifluoride etherate in dioxane. Reaction of the latter with tosyl chloride in pyridine gave the tosyl ester 231, which when treated with potassium I-butoxide in f-butyl alcohol/pyridine led to the orthoformyl ester 232 of angustmycin A. Careful hydrolysis of 232 with warm aqueous acetic acid and subsequent treatment with alcoholic ammonia to remove the acetyl groups gave angustmycin A identical with an authentic specimen.
E/ ‘i 3:
(1
u
\z
0 v 00
J:
7
V
z
(1
8 9
396
OH
239
242
Scheme 28
OCOC,,H,
BzCW
I
NHBz
OCOCeHS
240
BzOCH,
191
OH
Hot(
I
NH,
BzOCH,
241
OCOCGH,
Br
+ 69%
qcH3z qcH3 -
HOCH,
398
The Total Synthesis of Antibiotics
(3) Cordycepin
Cordycepin or 3’-deoxyadensosine was synthesized by Dalton and his associates.82 The protected deoxyribofuranosyl bromide 241 needed for this synthesis was prepared in part according to the procedure of Bakers3 beginning with the readily available 1,2-O-isopropylidene-D-xylofuranose233. Treatment of 233 with excess methyl chloroformate in pyridine afforded almost exclusively the monoester 234. Mesylation of 234 produced 235, which when subjected to acetolysis in acetic acid/acetic anhydride containing 1 % sulfuric acid generated a mixture of anomers (236). Methanolysis of 236 with 1 % methanolic hydrogen chloride gave the corresponding a- and pmethyl glycosides (237), which under the influence of methanolic sodium methoxide yielded the anhydrosugars. These anomers were easily separated by distillation and the /?-glycoside 238 was then hydrogenated over Raney nickel. This proceeded almost exclusively to methyl 3-deoxy-~-~-ribofuranoside (239), which was benzoylated to give 240. Reaction of 240 with 30% hydrogen bromide in acetic acid for a few minutes then gave the crude bromide 241, which was used directly in a condensation step with chloromercuri-6-benzoylaminopurine,which produced the protected nucleoside 242. Finally debenzoylation was achieved by boiling methanolic sodium methoxide to give cordycepin (191).
B. Pyrrolopyrimidine Nucleosides (Tubercidin, Toyocamycin and Sangivamycin) The only known naturally occurring pyrrolopyrimidine nucleosides have all been synthesized by Townsend and his The synthetic scheme followed is shown in Scheme 29. A previous attempt to synthesize toyocamycin (195) had failedss when no method could be found to desulfurize the penultimate compound 243. The
HO OH 243
244
245
246
193a Scheme 29
399
400
The Total Synthesis of Antibiotics
new approach started with 2-amino-5-bromo-3,4-dicyanopyrrole 244, which when refluxed with formamidine acetate in 2-ethoxyethanol afforded 245. Acetylation of 245 led to 246 and the latter was then fused with I ,2,3,4tetra-O-acetyl-~-~-ribofuranose in the presence of a catalytic amount of bis(p-nitrophenyl)phosphate8* to give the tetraacetyl nucleoside 247. Removal of the acetyl groups was accomplished by methanolic ammonia and yielded 248, which was dehalogenated to toyocamycin, identical with the natural product, by catalytic reduction in the presence of a palladium catalyst. Since hydrolysis of 195 previously had been shownsP to give the acid 249 (albeit in low yield), which underwent d e c a r b ~ x y l a t i o nwhen ~~ heated briefly at 238", this constitutes a total synthesis of tubercidin (193a) also. Sangivamycin (194) was in turn easily synthesized from toyocamycin (195) by oxidation with 30% hydrogen peroxide in concentrated ammonium hydroxide solution, thus completing total synthesis in this area. (Although several other subsidiary procedures also were reporteda4 for the syntheses of these compounds, the methods just described appear to be the best, overall.) C. Miscellaneous Nucleosides (Polyoxins, Blasticidin S)
Synthesis in the area of the polyoxins (e.g., 196) has been limited so far to the sugar moiety of which a suitably protected derivative (258) has been prepared" (Scheme 30) starting with 1,2: 5,6-di-O-isopropylidene-a-~allofuranose (250). The latter material, easily synthesizeds8 from D-glucose, was benzoylated and then treated with 1 . 5 % sulfuric acid in aqueous ethanol to give 251 by selective hydrolysis of the 5,6-isopropylidene group. Mesylation of 251 followed by boiling the product in dimethyl formamide with sodium benzoate afforded the desired material 252 in which a reversal of configuration at the 5-position had been achieved. Dcbenzoylation of 252 by means of sodium methoxidc in methanol, then treatment of the product with dimethoxypropane, yielded 253. Benzoylation followed by selective hydrolysis led to 254, which was in turn tritylated and mesylated to give 255. The amino group desired at the 5-position was now achieved in two steps first by boiling 255 with sodium azide in dimethylformamide, which afforded 256 (configurational inversion at C-5), then catalytic reduction over a palladium catalyst to give the amine, which was benzoylated immediately then detritylated with y-toluenesulfonic acid in acetone. The product of these transformations, 257, was next oxidized with potassium permanganate in acetone acetic acid solution and the resulting acid esterified with diazomethane to give 258. This protected derivative of 5-amino-5-deoxy-a-~-allofuranuronic acid, the sugar component of the polyoxins, is a suitable staging compound for further work toward the total synthesis of the polyoxins.
5. Nucleoslde Antibiotics
co45%gM Eo 401
55 %
70 %
BzO O l C M c ,
250
251
_ j
___f
__f
Bo Tho
BzO O l C M e ,
HO O-kMe,
BzO OACMe2 252
253
254
---+ *s%
l3zO O l C M e ,
BzO OJCMe2
255
256
50 % __+
BzO O l C M e , 257
258
Scheme 30
The carbohydrate component 268 of blasticidin S (197a) also has been synthesizeds9recently, by the same group who synthesized both the nucleoside portiong0 ("C"-substance) of gougerotin (197b) and a nucleoside9On corresponding to that of blasticidin S in which there is a hydroxymethyl group in place of the carboxylic acid. The synthesis of 268 is reported here because of its potential for the ultimate synthesis of blasticidin S itself. Of
The Total Synthesis of Antibiotics
402
the two methods described by Fox and his co-workerseefor the synthesis of 268, that shown in Scheme 31 appears to be the better.
H
0
5
47%lo260b
-
RO
B Z 0 3 '
OCH 3
80%
OCH3 260a R = H 260b R = M s
259
CH,OI%z
CHIOTr
CH,OTr __+
263
262
OCH3
OCH3
266
26% R = H 265b R = CH,
264
NH,
OCH
OCH
OCH, 261
60% (3180)
+ocH3 COiCHS
___t
267
Schema 31
RNH
CO, R'
268a R = R ' = H 268b R = Ac; R' = CH,
Methyl cc-D-galactopyranoside (259) was benzoylated selectively at the equatorial hydroxyl groups to give 260a according to the method of Reist et aLuoband then mesylated at the 4-position. The product 260b was treated with sodium azide in hexamethylphosphoramide and gave 261, which was
5. Nucleoside Antibiotics
403
debenzoylated with sodium methoxide in methanol then tritylated to give 262. Mesylation of 262 afforded 263, which when treated with sodium methoxide gave an epoxide. Detritylation of the latter led to 264 and oxidation of this material by potassium permanganate yielded the uronic acid 265a. The methyl ester 265b of this acid upon treatment with sodium iodide produced a mixture of iodohydrins that were mesylated in pyridine directly. Two products, which could be separated chromatographically, resulted from this reaction. One was the desired unsaturated azide 266 and the other an iodo mesylate that could be converted to 266 by a combination of tetramethylammonium chloride and zinc dust in pyridine. Reduction of 266 with sodium dithionite at pH 7 yielded the amino ester 267, which without purification was saponified by methanolic sodium hydroxide to give crystalline methyl 4-amino-2,3,4-trideoxy-a-~-erythrohex-2-enpyranosiduronic acid (268a), the methyl glycoside of the carbohydrate moiety of blasticidin S. Hydrogenation of 266 was also carried out using a platinum catalyst. Acetylation of the crude product then afforded 268b in almost quantitative yield. Theoretically all that remains to complete the total synthesis of this antibiotic is to couple a suitable derivative such as the pyranosyl chloride from 2681, with cytosine (or a derivative) at the 1-position of the latter to give the cytosinine (269a) (after deblocking), since blasticidin S has already been recon~tructed~~ (Scheme 32) from 269a and blastidic acid (270a). In NH,
N9 I
I
NHR'
I + R'NH c!! NCH,CH,CHCH,CO,H
oAN/
RO,C
NH
il
I
--+
CH3
:
NH? 269a R = H 269b R = CH3
270a R = H 270b R = C7H70CO-
27111 R = CH,; R' = C7H70CO197a R'= R' = H Scheme 32
404
The Total Synthesis of Antibiotics
this partial synthesis the dihydrochloride of 26913 and N,N'-dicarbobenzyloxyblastidic acid (270b) were coupled using N,N'-dicyclohexylcarbodiimide in methanol-acetonitrile containing triethylamine to give the protected derivative 271a. It should be noted that the peptide bond formation occurs preferentially at the desired site because the latter is the most basic of those available. Removal of the carbobenzyloxy groups was then accomplished in dry methanolic hydrogen chloride. Saponification of the resulting methyl ester on a basic ion exchange resin afforded blasticidin S. 6. PEPTIDE AND DEPSIPEPTIDE ANTIBIOTICS
Many antibiotic substances are small polypeptides and those of known structure range from about 15 (gramicidin A) down to 2 (penicillin) in amino acid residues. Others of unknown structure obviously contain many more peptide linkages since molecular weights in this class range up to 14,000 (saramycetins2). Most, however, are less than 2000. Only a limited number of the 50 or more identified compounds are of known structure and only a few are of therapeutic value. Of these we may include (apart from the penicillins) the following: the bacitracins (topical use); actinomycin D (cancer therapy); capreomycin and vioinycin (tuberculosis); the gramicidins and tyrocidins (topical use); the polymixins (urinary tract infections); the pristinamycins and staphylomycins (systemic infections). Total syntheses have been achieved in the areas of the gramicidins, the tyrocidines, the polymixins, and the actinomycins (depsipeptides). The individual classes are discussed in this section. Much synthetic work has been carried out by Kanekoe3 in an attempt to synthesize bacitracin A (272), but so far no success has been recorded. This appears to be due in part to the
I I I
I I
1:;-I
t
.ClI
tI 1.-i-LCU
t
272 (broken line represents link in cyclo form)
6. Peptide and Depsipeptide Antibiotics
405
few ambiguous points still unresolved about the s t r ~ c t u r e and ~ ~ *in~ part ~ to the racemization problems that are encountered when attempts are made to forge the peptide link between the thiazoline moiety and the L-leucine residue.OS Staphylomycin S (273) comprises only 5 % of the crude staphylomycin complex. It is an unusual substance both in its amino acid content, which
I I
CHCH,
I co
D
I
NCH,
0 I
273
L
CO-CH-NHCOCH-N-
I
C,H,
I
L
I
CH,
I co-
I I
CH,
CH-CH,C,H, L
CH,
includes 4-piperidene-Zcarboxylic acid, and in having one ester link in the macrocyclic ring. It is usually classified as a depsipeptide. Even more unusual in structure is its sister compound staphylomycin Mge(274),which constitutes
75 % of the staphylomycin complex. It is also identical with pristinamycin IIA and ostreogrycin and is again classified as a depsipeptide. Total synthesis of either 273 or 274 has not been reported. The structures of capreomycin and viomycin are not completely clears7 and total synthesis is still out of the question. Virtually none of the true macrocyclic depsipeptides is used therapeutically. Among the better known are the enniatinsgs-s9A (274s),B (274b), and C
406
The Total Synthesis of AntibloIics
L
I
D
L
CH3 D
274a R 274b R 274~ R 276 R
CH, L
l
D
= CH(CH3)qH,
= CH(CH3), = CH&H(CH3)2 = CH2C,H5
CH2(CHZ)6CH3 CHzoH I
I
HNCHC02CHCH2C0
I
275
I
COCH2CH0&CHN H
[
I
I
CH,(CH,),CH,
CH3
CH(CH3)2 CH(CH3)2
-0CHCONHCH-COZ I L
CHZOH
L
LHCONH-C-CO-I D
1,
CH(CH3)2 D
277
Valinomycin, which is the largest of these depsipeptides, contains a 36membered ringlo4 and is too toxic for use in mammalian species. It has in common with a number of these compounds the ability to complex with alkali metal cations and induce cation permeability of artificial and biological semipermeable membranes. Most of the synthetic work in this area has been carried out by Shemyakin and his group at the Academy of Sciences in Moscow. A reviewlosof much of the chemistry of the depsipeptides has appeared recently. Syntheses of the depsipeptides, as with the macrocyclic and linear peptides discussed in this section, follow well-established patterns. In the area of the depsipeptides (apart from actinomycin S ) , the total syntheses of the antibiotic beauvaricin (276) and serratomolide (275) are discussed since they illustrate the two different approaches possible in this area. A. Tyrothricin
The complex obtained by the extraction of acidified cultures of Bacillus breuis was termed “tyrothricin.”1o8This material is easily resolved into two fractions,
6. Peptide and Depsipeptide Antibiotics
407
one of which is soluble in ether or acetone and is termed the "gramicidin" fraction constituting 20-40 % of the initial mixture. The insoluble fraction is the "tyrocidine" fraction. The neutral gramicidin fraction*07was fractionated1Oe by countercurrent distribution and yielded a number of fractions, among which were three crystalline compounds, gramicidins A, B, and C. Although gramicidin A was thought to be cyclic originally, Witkop1OBJl0has shown that it is in fact a mixture of two compounds Val1-gramicidin A (278) and ]leu'-gramicidin A (279). The former is a linear decapeptide formylated
OCH-L-Val-Gly-L-Ala-D-Leu-L-Ala-D-Val-L-Val-~-ValL-Try-D-Leu-L-Try-D-Leu-L-Try-D-Leu-L-Try-NHCH,CH,OH 278
at the NH,-terminal residue and bearing ethanolamine at the COzH-terminal residue. In Ileul-gramicidin A (279) the terminal valyl residue at the beginning of the chain is replaced by an isoleucine residue. Both have been synthesized by Witkop.lll Gramicidin B again was shownl12 to be a mixture Val'gramicidin B and Ileu-gramicidin B. These are very similar to the gramicidins A except that the L-TryI2 residue is replaced by an L-phenylalanine. In gramicidin C one tryptophane residue is replaced by tyrosine. Gramicidin S (279) was isolated crystalline from Bacillus brevis by Gauseand B r a z h n i k 0 ~ a . lThe ~ ~ structure of gramicidin S was determined by Synge, Consden et aI.ll4 and its synthesis was accomplished by Schwyzer and Sieber.'l6 L-Val -+ L-Om + L-Leu
-3
D-Phe + L-Pro
L-Pro
t
L-Om + L-Val
1
t o-Phe t
L-Leu
1
279
The so-called gramicidins J1 and J,, which were originally thought to be cyclohepta- and cyclohexapolypeptides, recently have been shown to be identical"" with gramicidin S. The tyrocidine fraction of tyrothricin was separated by countercurrent distribution into three components, tyrocidines A, B, and C. Their structures were determined as 280,"' 281a,11* and 281b,'le respectively. Syntheses of tyrocidines A,12oB,lal and ClSahave been accomplished by a polypeptide antibiotic Izumiya and his group, who have also tyrocidine E, recently isolated by Kurahashi et al.,124 from a culture of Bacillus breuis ATCC8185. This is the same as tyrocidine A except that the L-Tyr residue of the latter is replaced by L-Phe.
408
The Total Synthesis of Antibiotics
L-Val
----f
L-Orn --+L-Leu -+ D-Phe N H2
NHZ t
1
L-Val
L-Tyr
--f
-----f
I
L - A s ~t D-Phe
L-Orn -+ L-Leu NH2
-----f
D-Phe
NHZ
I I t L-GIu t L - A s ~t 281a [XI= D-Phe 281b [X)= D-Try
1
L-Pro 280
f-
L-Phe
---f
L-Pro
I
[XI +-- L-Try
In the area of the tyrothricin antibiotics the syntheses of Val1-gramicidin A, gramicidin S, and tyrocidine A are representative and are thus discussed individually. In these syntheses the following abbreviations are used : Z = carbobenzyloxy ; Np = p-nitrophenyl; BOC = r-butyloxycarbonyl; Tos = p-methyltoluenesulfonyl ; pmz = p-methoxy carbobenzyloxy and PHT = phthalyl. Vull-grumicidin A (278)
The total synthesis of this molecule was carried out by Sarges and Witkop.”’ All of the amino acids in this molecule are neutral amino acids, so no particular difficulties were expected in their construction. The acid sensitivity of the tryptophane moiety, however, restricted the procedures that could be used. For this reason Sarges and Witkop planned to synthesize and join together the octapeptide Z-~-Val-Gly-~-Ala-~-Leu-~-A~a-~-Val-~-Val-~ R 282 and the heptapeptide ethanolamide L-(H)-Try-D-Leu-L-Try-D-Leu-LTry-D-Leu-L-Try-NHCH,CH,OH 283, either by the azide method or by coupling with dicyclohexylcarbodiimide at low temperature, methods that would largely avoid the bane of all such syntheses uiz. racemization. The heptapeptide 283 was prepared by stepwise synthesis from the carboxyl end by the mixed anhydride method. This method (the general case is shown below) is fast and convenient and minimizes racemization. It was also used for the synthesis of the octapeptide 282 (R = OCH,), but unfortunately the ester function could not be saponified to give the desired fragment 282
+ +
ZNHCH(R)CO,H CIC0,Et % ZNHCH(R)CO,OCOEt ZNHCH(R)C020COEt NH,CH(R’)COR” ZNHCH(R)CONHCH(R’)COR” NH,CH(R)CONHCH(R’)COR” (R = alkoxyl. or amine. etc.)
-+
...
6. Peptide and Depsipeptide Antibiotics
409
(R = OH) and accordingly an alternate strategy was adopted. The ester function was converted to the hydrazide and coupled via the azide with the heptapeptide ethanolamide 283. The product had the desired properties of the authentic N-carbobenzyloxydesformylgramicidin A but at the time simply because of lack of material, experimental conditions for the hydrogenolytic cleavage of the terminal carbobenzyloxy protecting group could not be found. Later it was found that the reaction could be accomplished in methanol over palladium black. In the meantime a more convenient approach was explored which consisted of coupling the pentapeptide derivative Z-L-Val-Gly-L-Ala-D-Leu-L-Ala-OH (284) with the decapeptide ethanolamide D-(H) Val-L-Val-D-Val-L-Try-DL~u-L-T~~-D-L~u-L-T~~-D-L~u-L-T~~-NHCH,CH,OH (285), which was prepared by the stepwise prolongation of the heptapeptide 283 (discussed earlier), again using the mixed anhydride method for the introduction of each amino acid residue. The pentapeptide 284 was also synthesized by this technique starting with methyl L-alaninate to give eventually the pentapeptide methyl ester which was saponified without difficulty. When 284 and 285 were coupled under the influence of dicyclohexylcarbodiimide the expected N-carbobenzyloxydesformylgramicidin A 286 was obtained. 286 287 288 278
Z-L-Val-Gly . . . D-L~u-L-T~~-NHCH,CH,OH L-(H)Val-Gly . . . D-Leu-L-Try-NHCH,CH,OH OCH-L-Val-Gly . . . D-L~u-L-T~~-NHCH,CH,OCHO OHC-L-Val-Gly . . . D-L~u-L-T~~-NHCH,CH,OH
Hydrogenation of 286 over palladium black in methanol afforded the desformyl polypeptide 287, which when treated with formyl acetic anhydride led to O-formyl Val1-gramicidin (288). Saponification of the latter with 1 N sodium hydroxide then gave the antibiotic itself identical in all biological and chemical respects with the naturally occurring material. In a completely analogous fashion Z-L-Ileu-Gly-L-Ala-D-Leu-L-Ala-OH was coupled with the decapeptide 285 to give, after the three final steps, Ileul-gramicidin A. Gvamicidin S (279)
Although the linear sequence of gramicidin S had been synthesized in a number of laboratories, Schwyzer and Sieler115were the first to prepare the cyclic form. In fact this was the first synthesis of a cyclic peptide antibiotic and it succeeded only because of the suitable cyclization methods developed by these workers. Two syntheses were devised, both of which depended on the synthesis of the protected pentapeptide Z-L-Val-L-Om (Tos)-L-Leu-DPhek-Pro-OCH, (289). This was carried out by the method of activated esters (Scheme 33). The coupling of Z-L-Leu-ONp and H-D-Phe-OC,H, was
e
3
410
X
I
411
412
The Total Synthesis of Antibiotics
carried out 20" in tetrahydrofuran. The product was saponified to give Z-L-Leu-D-Phe-OH, which was converted to the corresponding cyanomethyl ester by means of chloroacetonitrile in the presence of triethylamine. Condensation of this active ester with L-proline methyl ester then afforded the tripeptide which was hydrogenolyzed to give H-L-Leu-D-Phe-L-Pro-OCH, suitable for further coupling. The dipeptide methyl ester Z-L-Val-L-Om (Tos)-OCH, was prepared in the same way then converted to the hydrazide and subsequently to the azide Z-L-Val-L-Orn(Tos)-N,, which when condensed with the tripeptide prepared above afforded the desired protected pentapeptide 289.Catalylic reduction then removed the carbobenzyoxy group and the methyl ester 290 thus obtained was N-tritylated to give the N-trityl ester 291. Saponification of 291 in a large excess of alkali then produced the N-protected peptidic acid 292, from which two syntheses of gramicidin S were developed. In 292 was coupled with the amino ester 290 by The resulting means of l-cyclohexyl-3 [2-morpholinyl-(4)-ethyl]carbodiimide. N-trityl decapeptide ester was saponified to the corresponding acid, which was then converted to the p-nitrophenyl ester by means of p-nitrophenyl sulfite in pyridine. Detritylation was accomplished by treatment with trifluoroacetic acid at -5' for 15 minutes and resulted in the trifluoroacetate of the decapeptide p-nitrophenyl ester. Cyclization was brought about by the dropwise addition of a dimethylformamide solution (containing a little acetic acid) of this compound to pyridine using a high dilution technique (3 x 10-3M solution). Subsequent purification of the cyclic peptide afforded a 28 % yield of crystalline material (293)identical with the bis-tosyl derivative of natural gramicidin S. The tosyl groups which had been used to protect the 8-amino groups of the ornithine residues were now removed by treatment of 293 with sodium in liquid ammonia. The product (70-90% yield) was identical in all respects with native gramicidin S (279). In the second procedurelZ5for the synthesis of 279 the N-tritylpentapeptide 292 was esterified by means of p-nitrophenyl sulfite then detritylated. Cyclization of the resulting amino ester then occurred with doubling to give 293 directly. This type of twinning was not unknown in other cases but linear pentapeptides do not always cyclize exclusively with doubling. However, in this case no cyclosemigramicidin could be detected, so the transition state for the formation of the cyclodecapeptide must be very favorable. Tyrocidine A (280)
In the synthesis of tyrocidine A , Izumiya and his associates followed a pattern not dissimilar to that used by Schwyzer for gramicidin S. I n general two pentapeptides were constructed and subsequently linked to give a decapeptide. Since Izumiya intended to cyclize the decapeptide by the use of the active p-nitrophenyl ester method of Schwyzer, he was guided in his
6. Peptide and Depsipeptide Antibiotics
413
choice of the two peptides by the limitations of the technique. This relates to the racemization that can occur at the amino acid moiety during .ester formation, except when the amino acid residue is glycyl or prolyl. For this reason he chose to synthesize a decapeptide having at its terminus a proline group. The grid for the total synthesis of tyrocidine A is shown in Scheme 34. It needs little comment except to say that the isolated amino group of the ornithine moiety was protected up to the penultimate step as the carbobenzyloxy derivative as opposed to the tosyl derivative used by Schwyzer. This necessitated an alternate method of protecting the amino group end of the peptide chains during their synthesis. In the case of the pentapeptide L-Phe-D-Phe-L-Asp-L-Glu-L-Tyr-OH this was done using the p-methoxycarbobenzyloxy group, which can be removed by trifluoroacetic acid even in the presence of the sensitivep-nitrophenyl group. In the case of the pentapeptide L-Val-L-Orn-L-Leu-D-Phe-D-Pro-OH it was accomplished by means of a r-butyloxycarbonyl function which was eventually removed by hydrogen chloride in ethyl acetate. The couplings of Z-D-Phe-OH with H-L-Pro-OC,H, and of Z-L-Leu-OH with H-D-Phe-L-Pro-OC,H, and of pmZ-L-Phe-OH with H-D-Phe-OEt were all carried out by the mixed anhydride method. The dipeptide BOC-L-Val-L-Orn(Z)OC,H, was made by dicyclohexylcarbodiimide coupling of BOC-L-Val-OH and H-L-Orn(Z)OEt using conventional procedures. It is interesting to note that the first attempt to synthesize the decapeptide 296 was made by coupling the azide with the ester derived from 296 after removal of the BOC group. However, the Japanese workers were unable to hydrolyze the ester function in the coupled product-a result that parallels Witkop’s experience”’ with the octapeptide 282 (R = OCH,) discussed earlier. The synthesis scheme that they finally adopted produced pure crystalline tyrocidine A hydrochloride identical with the same salt of the natural material. The syntheses of the remaining members of this group were accomplished after the same fashion. B. The Polymixins
The polymixins form a group of closely related cyclopolypeptides that are obtained from various strains of Bacillus pofymyxa. They are potent antibiotics against gram-negative organisms but are somewhat limited by their toxicity. Structure determination in this area was for some time fraught with and only after considerable synthetic and degradative work were the questions of peptide sequence and ring size solved. The polymixins have the general structure 297 and are classified as A (297a), B (297b),B, (297c), C (297d), D, (297e), D, (297f), and M (297g). In polymixin the central Dab of the three Dab moieties in the ring has been
N
N
4
-
414
N
N
X
2 0
-
N
N
f
N
%
N
k
N
k
X
Y
Y
415
416
The Total Synthesis of Antibiotics
exchanged for D-Leu. The circulins and the colistins also belong to this family. Polymixin E is identical with colistin A. L-Dab R-L-Dab
-+ L-Thr -+ W
297
B1 (297b)
Bz (297c) C D1 Dz M
(2973) (297e) (297f) (2978)
--t
t t L-Thr
----t
D-X --+ L-Y
L-Dab
+ L-Dab
t- L-Dab
Dab = u,y-diaminobutyric acid
R W (+) CH3CHzCH(CH3)(CHz)4C0 L-Dab CH&H(CH3)(CH&CO (+) CH3CHZCH(CH3)(CHz),CO (+)CH3CHzCH(CH3)(CH2)4C0 CH3CH(CH,)(CH.J4CO (+) CH,CH,CH(CH3)(CH,)4C0
L-Dab
X D-Phe D-Phe
?
?
D-Ser D-Ser L-Dab
D-Leu D-Leu L-Thr
Y L-Leu L-Leu ? L-Thr L-Thr L-Dab
Only the structures just mentioned are known with certainty. One compound of this group has been synthesi~ed,'~'polymixin B,. A number of isomers of B, that do not occur naturally are also known. The synthesis of 297b was accomplished by synthesizing the protected polypeptide moieties 298, 299 and 300, coupling them as discussed below, then cyclizing the heptapeptide chain after removal of the blocking groups. The syntheses of these peptides are shown in Scheme 35. Standard methods were used to couple the amino acids in most of the steps and are obvious from the schemes. The introduction of the (+)-6-methyloctaroic acid (MOA) into the tetrapeptide 299 was carried out by causing the acid to react with H-~-Dab(Nz)-~-Thr-ocH, under the influence of carbonyldiimidazole. The necessary dipeptide 300 was obtained by condensing N"-phthaloyl-N"-carbobenzyloxy-u,y-diaminobutyricacid with D-phenylalanine r-butyl ester in the presence of dicyclohexylcarbodiimide. The phthaloyl group was then removed by means of hydrazine. The coupling of the synthesized pepides was carried out as follows. The y-BOC group of 299 was removed by means of trifluoroacetic acid and the resulting amine was condensed with the azide derived from 298. The octapeptide that was produced was converted to the corresponding hydrazide and then to the azide which was coupled with the dipeptide 300 to give the
y-z
I
298
y-z
I
-
299
MOA
L-Dab-y-Z
I I
I
L - T ~
BOC-L-Leu-L-Dab-L-DabL-ThrNHNH, - - y-BOC-L-Dab-OCH, -
- - H-L-Dab-D-Phe-OBu'
BOC --OH BOC--
OPN
H -- OCH,
BOC
OCH,
H
OCH,
MOA
OCH,
MOA
NHNH,
OCH,
Z --OH
MOA
OCH,
Z -- OCH,
MOA
NHNH,
H -- OCH,
H--
OCH,
MOA
BOC -- ONP
BOC-BOC--OH BOC--
ONP
ONP
H -- OCH,
BOC
OCH,
H
OCH,
BOC
OCH,
H
OCH,
BOC
OCH,
BOC
NHNH,
6. Peptide and Depsipeptide Antibiotics
419
branched decapeptide 301. The terminal t-butyloxycarbonyl and i-butyl
BOC-L-Leu-L-Dab-L-Dab-L-Thr-L-Dab-L-Dab-D-Phe-OBu
I
y-z
I
y-z
/I
I
Y-z L-Dab-y-Z
I I L-Dab I
L-Thr
MOA 301
groups were removed by trifluoroacetic acid at room temperature and the polypeptide amino acid that resulted was cyclized by means of dicyclohexylcarbodiimide in dimethylformamide-dioxane at high dilution. Finally the carbobenzyloxy groups were removed by cleavage with sodium in liquid ammonia. The product from this reaction after purification had all the physical and biological properties of polymixin B,. C.
Cyclic Depsipeptides
Although actinomycin D contains only one ester linkage in the polypeptide half of the molecule and has also a large heterocyclic nucleus, it is still considered a depsipepptide and is included in this section along with the more authentic cases, serratamolide and beauvericin.
Actinomycin D The actinomycins were first discovered by Waksman and Woodruff127who isolated actinomycin A from a species of Streptomyces antibioticus. The isolation of the various components of the complex, their structure elucidation, and their chemistry have been reviewed by Brockmann.12*The major component of the complex is actinomycin D or C, (302). The first synthesis of the compound was reported by Brockmann12vin 1964 and is shown in Scheme 36. The tetrapeptide 304 required for subsequent condensation was prepared by a series of carbodiimide condensations starting with Nformylvaline and the benzyl ester of L-proline. The resulting dipeptide was hydrogenated to cleave the benzyl group and the acid obtained used in a subsequent condensation. The formyl group of the tetrapeptide was removed by hydrolysis with a small amount of aqueous hydrochloric acid in benzyl alcohol, thus minimizing ester hydrolysis and allowing the formation of 304 (as its hydrochloride) in good yield. The other component 305 was
L-MeVal-OBz
I Sar I
L-Pro
I D-Val I
90% __f
I Sar I
+
I
OBz CH3 305
304
L-MeVal-OBz
ae%_
L-Pro
(HI
303
L-MeVaI-OH
I
Sar
I I
L-Pro
+
D-Val
I L-Thr I
-
80 % from 356
CH,
GNH2
306
(not isolated)
C-0
CH 3 OH
307
Scheme 36
420
I
@ON02
D-Val
Formyl
I Sar I L-Pro I D-Val I L-Thr I
L-Thr (OH1
L-MeVal-OBz
L-MeVal-OH
L-MeVal-OH
I
I Sar
Sar
I
I I
L-Pro
L-Pro
I
D-Val
HO-C-CH(L) NH
NH
I
I
C=O
CH3
CH3 308
L-MeVal
L-MeVal
L-Pro
L-Pro
D-Val
D-Val
0
I
I
c=o
LCH--LH(L)
I
I
CH, N H
I
o=c
‘
C==O
0
I
I I
HN CH,
I c=o
302
Scheme 36 (Continued) 421
422
The Total Synthesis of Antibiotics
prepared by allowing 2-nitro-3-benzyloxy-4-methylbenzoylchloride to react with the sodium salt of threonine.laoThe condensation of 304 with 305 was accomplished by the use of N-ethyl-5-phenylisoxazolium-3’-sulfonate (Woodward’s reagent) in nitromethane in the presence of triethylamine. The resulting compound (306) was submitted to catalytic hydrogenation to remove the two benzyl groups. Without characterization of the intermediate phenolic acid 307, it was oxidized with potassium ferricyanide solution at pH 7.2 in an oxidative coupling reaction which was known in the case of simpler o-aminophenols. Actinomycin D (302)was now obtained by treating 308 with a mixture of acetyl chloride and N,N’-carbonyldiimidazolein tetrahydrofuran solution followed by chromatography of the crude product. The synthetic material was congruent with the natural product. More recently a new synthesis of actinomycin D has appearedl3I which permits its preparation, and that of analogous substances, in substantial quantity. In this synthesis the difficult final lactone ring closure step of the Brockman synthesis is avoided and formation of a peptide linkage becomes the last reaction of the sequence. The key reaction, the cyclization to the pentapeptide lactone, was carried out by peptide bond formation between the prolyl and sarcosyl residues of 308a using the p-nitrophenyl active ester
I
0-Me.Val-Sar-H ~O-L-Tlir-D-Val-L-Pro-oNp
308a
technique. The ester bond between the carboxyl group of N-methylvalyl and the hydroxyl group of the threonyl residues was formed by a reaction of r-butyloxycarbonyl-L-threonineand the mixed anhydride from benzyloxycarbonyl-L-N-methy haline and isobutyl chloroformate. Beauvericin (276) The synthesis132of beauvericin is typical of this class of depsipeptide where peptide and ester functions alternate. The synthetic grid is shown in Scheme 37 and the methods are relatively simple because in this case only two residues are used to build the molecule, L-N-methylphenylalanine and D-2-hydroxyisovaleric acid. N-Carbobenzyloxy-N-methylphenylalaninewas coupled with 1-butyl D-a-hydroxyvalerate by means of carbonyldiimidazole. The product, 309 was in separate experiments (a) treated with trifluoroacetic
hl
423
424
The Total Synthesis of Antibiotics
acid and (b) hydrogenated catalytically to give 310 and 311, respectively, the two components of the next condensation. This condensation was accomplished by treating 310 with thionyl chloride in triethylamine and subsequently adding 311. In a completely analogous fashion 312 was converted into 313. The latter was treated with hydrogen bromide in acetic acid to remove the terminal protecting groups and the product was subjected to the action of thionyl chloride in triethylamine at high dilution. The synthetic material was essentially identical with the natural product. In each step of this synthesis the yield was better than 70% except the last which was 25%. Serratamolide (275)
In his synthesis ofthe 0,O'-diacetyl derivative of the antibiotic serratamolidelO1 Shemyakin13*"very cleverly made use of the known hydroxyacyl insertion reaction:
CH,OBz
I
(PHT ) N ~ H C O , C H
CH,OBz
I L-(PHT)NCHCO,H
,/
\
315
CH20Bz
I
N HzCHC02CHB
(PHT)NCHCONHC'HCO,CH,
318
0
CH,OR
319
R R
317
= Bz = Ac
6. Peptide and Dipsipeptide Antibiotics
425
For this synthesis D-p-benzyloxydecanoic acid and the diketopiperazine 319 were required. The latter was synthesized as follows: phthalylation of 0-benzyl-DL-serine afforded N-phthalyl-0-benzyl-DL-serine,which was then resolved to give the L-isomer 314. Methylation with diazomethane gave 315, which was hydrazinolyzed in good yield to methyl 0-benzyl-L-serine (316) isolated as the hydrochloride. Condensation of 314 with 316 in the presence of dicyclohexylcarbodiimide afforded the peptide 317, which on treatment with hydrazine hydrate led to the dibenzyl ether 318. Hydrogenolysis and acetylation then afforded the desired 319. D-p-benzyloxydecanoic acid was prepared from the corresponding acid, silver oxide, and benzyl bromide. The acid chloride 320 when heated in boiling toluene with 319 led to 321, ~H~OAC
OBz
+ CH,OAc 319
321
426
The Total Synthesis of Antibiotics
which was then hydrogenolyzed and afforded, via 322, the 0,O'-diacetylderivative 323 of serratamolide identical with that derived from the natural product. In an attempt to synthesize serratamolide itself (275), the diketopiperazine 324 was prepared. However, hydrogenolysis of the four benzyl groups did
325
OH
not give the desired material but only the product (325) of N -% 0 acyl migration. For the hydroxyacyl insertion reaction to work in this system the hydroxyl groups of the diketopiperazine moiety must be protected. 7.
MACROLIDE ANTIBIOTICS
The complex macrolide antibiotics can be subdivided into three classes based on both biological activity and structural features. The members of the first ~ ~ are typical group, of which e r y t h r ~ m y c i n l(326) ~ ~ and c h a l ~ o m y c i n '(327)
326
H
7. Maerolide Antibiotics
427
327
examples, have a biological activity similar to that of penicillin and have been used extensively in the past because of the high incidence of penicillinresistant infections. Their importance has diminished recently with the availability of penicillins not affected by penicillinase. The second group comprises such substances as f i l i ~ i n '(328), ~ ~ amphotericin B,I3O and nystatin.lg7 This group is noted for its antifungal activity
,
C H (0H 1C, H, 0 OH OH OH OH OH OH
328
and, although mainly used topically, individual members such as amphotericin B are used intravenously for fungal infections of the blood. In comparison with the first group they have much larger lactone rings, which always contain extensive conjugated unsaturation. Sugars may be attached to the lactone ring. None of the members of either group have been synthesized. The presence in both groups of a large number of contiguous asymmetric centers in an aliphatic chain presents almost insuperable problems to the chemist bent on stereoselective synthesis. Certainly it seems that at least one probable way in which the synthesis of the lactone moiety, of say erythromycin, will be achieved will be by the synthesis of a perhydro suitably substituted tricyclic intermediate, followed by cleavage of the transannular bonds. A third class of macrolide exists; by comparison with the two groups just mentioned the members are relatively simple in structure. Representatives of this class are curvularin (329), zearalenone (330), and r a d i c i ~ o l '(331). ~~
428
The Total Synthesis of Antibiotics
H0
HO 329
CH3
330
33 I
All of these compounds can be regarded as orsellinic acid derivatives generated without doubt from polyketide precursors. The syntheses of the di-omethyl e t h e 9 of 329 and of 330 itself have been reported. These are described only because they represent rare examples of the synthesis of naturally occurring macrolides and not because of any outstanding biological activity. A. Curvularin (329)
Initially, attempts were made to synthesize di-o-methylcurvularin (335) by the lactonization of 332. However, all attempts to cyclize the compound with
332
trifiuoroacetic anhydride or dicyclohexylcarbodiimide failed. Lactones of this specific type are difficult to hydrolyze and the corresponding hydroxy acids are resistant to lactonization. This appears to be due to the steric hindrance offered by the methyl group and in part perhaps to the conformation of the macrocyclic ring. An alternate approach was therefore taken, using a Friedel-Crafts reaction. This type of reaction had been used successfully in the past for the synthesis of several medium-sized ring ketones.'40
7. Maerolide Antibiotics
429
OCH,
I
+ CH,CH(CHJ6C02Bz I
CH ,O
%
OH 333 R = Bz 334
CH ,O
R
=H
5
CH Benzyl 7-hydroxyoctanoate, prepared by benzylation and sodium borohydride reduction of 7-oxooctanoic acid, reacted smoothly at room temperature with 3,5-dimethoxyphenylacetyl chloride to yield the diester 333. Hydrogenolysis afforded the acid 334, which was cyclized to 335 by allowing it to stand in a dilute solution of trifluoroacetic anhydride and trifluoroacetic acid. Attempts to demethylate 335 to curvularin were not recorded.
B. Zearalenone (330) The first synthesis of zearalenone was reported by Taub et al.141Zearalenone is produced by the fungus Gibberella zeae and has anabolic and uterotrophic activity. As a prelude to the synthesis of 330 the recyclization of the seco acid (336) of di-o-methylzearalenone (337) was studied, since it was visualized that this reaction would be the penultimate stage in the synthesis of zearalenone. This was accomplished in 80% conversion yield, by 1 molar equivalent of trifluoroacetic anhydride in very dilute benzene solution at 6" for 18 hours. That ether cleavage of 337 to zearalenone by boron tribromide could be accomplished was also ascertained at this point. Having secured this information attention was turned to the total synthesis of the seco acid itself. It was envisaged that this could be accomplished by connecting the two components 338 and 339. In 339 the potential functionalities of zearalenone at C-6 and C-10 are masked by internal ketal formation. This compound was synthesized starting from 5-oxohexanoic acid by reduction with sodium borohydride followed by acid treatment which afforded 340. The reaction of this tetrahydropyrone with 4-pentenyl magnesium bromide led to the keto acid 341, which was converted thermally to 342.
CH 3O
J5CEzL
CH30
0
336
337
CH30
-
Br
CHO
338
CH,CO(CH,),CO,H
OR
339
CH3
340
A+ CH3
dBH0 341
342
OCH
343
Br
__+
OCH3
-./--w 344
#3P+
OCH
OCH, 430
339
345
7. Macrolide Antibiotics
431
Methanolic hydrogen chloride then afforded 343, which was ozonized at -60" in methanol and the ozonide reduced with sodium borohydride to the carbinol 344. This carbinol was, however, extremely sensitive to acid, giving the spiran 346 with great ease. Under basic conditions then 344 was
converted to the desired bromide via the tosylate. Triphenylphosphine in hot methanol then transformed 339 into the salt 345. In both of the transformations from 344 to 345 the ketal group suffered some cleavage to the acyclic hydroxy ketone. This did not constitute a problem since the cyclic ketal was easily reformed in hot acidic methanol. In preliminary experiments it was found that whereas the ylide
+,P=CHCHS when condensed with the methyl ester of 338 gave complex results, condensation with the sodium salt of this acid in DMSO gave the desired product cleanly. When these conditions were used with the ylide from 345 condensation occurred smoothly to give a mixture of cis and trans seco acids 336 after work-up. Cyclization and demethylation under the previously mentioned conditions then afforded dl-zearalenone.
& + & C0,Na
CH30
$,P=CH
CHO
OCHS
OCH,
__f
CH ,O
OH
336
330
+
432
The Total Synthesis of Antibiotics
A second synthesis of this macrocycle has been reported142by a group at Syntex (Scheme 38). This again devolves on lactone formation in the penultimate step. The complete assembly of carbon atoms was smoothly elaborated by a Wittig reaction involving the ethyl ester of 338 with 347.
+
u u
CH 3O
347
348
349
CH,O
0 350
Scheme 38
This should be contrasted with the complex products obtained from the ethyl ester of 338 and the Wittig salt +,P+C-HCH, mentioned previously. Selective cleavage of the terminal ketal group was possible with 348 and the ketone so obtained was reduced by borohydride to 349. Base-catalyzed cyclization of 349 then led to 350, which when hydrolyzed to remove the ketal group afforded dl-zearalenone. Differentiation of the two ketone functions of 351, the diketone derived from 348, was also achieved in another way. Borohydride reduction of 351 led to 352, which when treated with sodium hydride afforded the ether 353.
7. Macrolide Antibiotics
433
Further treatment with the same reagent at elevated temperatures then gave the lactonic ether 354, which under no circumstances could be modified to give zearalenone. OCH,
2351
OCH,
COzEt
CH ,0 352
OCH,
__f
CH,O
CH
353
354
The foregoing methods for the synthesis of zearalenone use the third of the three conventional methods for the synthesis of macrolides. These (6) methods are (a) Baeyer-Villiget oxidation of macrocyclic peracid oxidation of bicyclic en01 ethers,"' and (c) the direct cyclization of hydroxy acids and esters. Studies in connection with the synthesis of zearalenone and curvularin by the second of these methods have been reported by Immer and Bagli. In their investigations a series of compounds of the type 355 were synthesized and oxidized by means of m-chloroperbenzoic acid to 356. No successful synthesis of zearalenone has yet emerged, although an
355
356
434
The Total Synthesis of Antibiotics
isomer of curvulatin has been prepared.lq5 Undoubtedly the presence of a double bond in the lactone ring of 355 will present complications to this method. 8. MISCELLANEOUS ANTIBIOTICS
There remains a group of antibiotics that in many instances are of very great therapeutic value but which do not fit into any one class and for convenience are simply assembled and discussed in this section. Of the 15 or so compounds that comprise this group only 6 have been synthesized, anthramycin (357), chloroamphenicol(358), cycloserine (359),cycloheximide (360), griseofulvin (361), and novobiocin (362). The synthesis of each of
CH20H
I
357
358
360
,OCH,
H 361
N 0.1
8. Miscellaneous Antibiotics
435
CH,
362
these molecules is discussed here with the exception of cycloserine whose preparation14"presented no special problems. Of the remaining compounds, coumermycin A,1P7 (363) [which is closely related to novobiocin (363)], d a ~ n o r n y c i n ' ~(364), ~ f ~ m a g i l l i n(365), ~~~ fusidic acid160(366), mitomycin C (367), monensinl6l (368), rifamycin BlSa (369), s t r e p t ~ n i g r i n(370), ~ ~ ~ and streptovitacin A16' (4e-hydroxycycloheximide), only fumagillin166~160 and mitomycin C have seen any serious attempts being made toward their synthesis. The significant synthetic work dealing with mitomycin C is discussed briefly in this section.
CH3 363
364
367
C H 3QC H 3 C H 3
368 436
8. Miscellaneous Antibiotics
437
369
OCH 3
cH30* HO
A. Antbramycin (357) The total synthesis15' of anthramycin, a potent antineoplastic agent, was carried out by the Hoffmann-LaRoche group who both isolated16* the compound and proposed its During the structure work it had been shown that anthramycin methyl ether (371) could be converted, via anhydroanthramycin, to anthamycin itself. Since 371 was well characterized and comparatively stable, the problem was reduced itself to the synthesis of this compound. As a primary synthetic objective the corresponding cyclic amide (372) was chosen. A particularly attractive approach to this material
371
372
438
The Total Synthesis of Antibiotics
was to use a naturally occurring amino acid, which would result in the final product being optically active rather than racemic. The synthesis (Scheme 39) began therefore with the acylation of L-hydroxyproline methyl ester (374) with 3-benzyloxy-4-methyl-Znitrobenzoylchloride (373).The product 375 was reduced with sodium dithionite to the corresponding amine, which when heated with aqueous hydrochloric acid cyclized to the lactam 376.Oxidation OC,H,
cH37y:;l 373
374
375
376
377
H c& * 3
__f
0
C'O.'('*Hj 378
OC',fl;
0 30 %
C t j 3 7 J 4 N+ /h C N 0 Scheme 39
379
from 428
__j
8. Miscellaneous Antibiotics
439
380
CH
372
Scheme 39 (Continued)
of 376 with chromic acid led to 377, which was condensed with the sodium salt of triethylphosphonoacetate at 0" to give as the major product the P,y-unsaturated ester 378. Reduction with diisobutylaluminum hydride at -60" gave a labile aldehyde which was immediately converted to the bisulfite adduct and then by treatment with potassium cyanide to a mixture of the epimeric c y a n o h y d r i n ~Treatment .~~~ of this product with methanesulfonyl chloride in pyridine followed by boiling the mesylates in benzene with triethylamine afforded a trans:cis::4: 1 mixture of the conjugated nitriles (380). These isomers could be separated by chromatography but were found to be easily interconvertible. For this reason the mixture was simply debenzylated by means of a combination of boron trifluoride etherate and trifluoroacetic acid at ambient temperatures. The resultant mixture of phenolic cis- and trans-nitriles was then heated in aqueous trifluoroacetic acid at reflux temperature and provided as the major product the desired intermediate 372. The remaining problem in the total synthesis consisted formally of specificallyeffectingreduction of the secondary amide group of 372.Attempted reduction with an excess of lithium aluminum hydride left 372 unchanged, so it was decided that a derivative would have to be prepared to obtain the desired result. It was envisaged that 381 might be suitable because simple N-methylphenylamides are converted in good yield to aldehydes by lithium aluminum hydride. Unfortunately the desired intermediate could not be prepared from 372.Nevertheless, the corresponding derivative 382 could be prepared from the trans-nitrile (itself obtained by debenzylation of 380) by condensation with benzaldehyde dimethylacetal at 200".This was converted to the amide 381 by hot polyphosphoric acid. Reduction of 381 with sodium
440
The Total Synthesis of Antibiotics
0
381 R = CONH, 382 R = CN
borohydride in methanol at 5' followed by hydrolysis of the carbinolamine with very dilute acid in methanol then afforded anthramycin methyl ether (381) in 70% yield identical in every respect with an authentic sample. A different approach to the anthramycin skeleton has recently been reported160 but as yet a second synthesis of the antibiotic itself has not appeared.
B.
Griseofulvin (361)
Griseofulvin was first discovered161in 1939, but its complete structure was not established162until 1952. Four different groups have completed a total synthesis of this molecule, three of them being announced initially in the same year (1960). Scott and his associates163examined experimentally the suggestion by Barton and CohenIs4 that griseofulvin was derived biogenetically from the benzophenone 383 which underwent C - 0 oxidative coupling to give dehydrogriseofulvin 385 via the diradical384. Enzymatic reduction of 385 would then H$20
OCH
CH30
CH 3OW CI
CH o3
"
-
383
OCH,
C H 3 0CI@ b CH, o 384
385
8. Miscellaneous Antibiotics
441
give 361. The desired benzophenone was eventually synthesized, after many attempts, by the Friedel-Crafts reaction of the acid chloride 387 with the phenol 386, followed by alkaline hydrolysis to remove the protecting methoxycarbonyl group. The overall yield was poor (-15 %), but the benzophenone
OCH,
OCH,
I
+ CH ,O *OH
,
CH 3
CI
OCOzCH 387
386
Cl
383
was available from griseofulvin itself by dehydrogenation with selenium dioxide to 385 followed by catalytic reduction over a platinum catalyst which simply caused hydrogenolysis of the C-0 spiro bond. Internal coupling of 383 to regenerate dehydrogriseofulvin 385 proceeded smoothly as predicted ( 5 0 4 0 % yield) in a mildly alkaline solution of potassium ferricyanide. Selective reduction of 385 to d-griseofulvin was then accomplished by reduction over a rhodium-charcoal catalyst poisoned by the addition of 3 % selenium dioxide. The yield was 30%, dropping to 2 % in the absence of the added poison. Resolution of the racemic material was accomplished by acidic hydrolysis to dl-griseofulvic acid 388, which was resolved as its quinine metho salt. The d-form of 388 had already been shown to regenerate
CH3?
OOH
Cl
CHJ 388
CH39
Cl
OOCH.
C-H, ‘H
361
griseofulvin (361) when treated with diazomethane. Thus a formal synthesis of griseofulvin had been completed. The synthetic pattern followed by the Merek group1s6for the synthesis of 361 paralleled that described by Scott and his associates1e3but was a much
442
The Total Synthesis of Antibiotics
more thorough investigation. They found that the benzophenone 383 was best produced by allowing the free acid 389 and the phenol 386 to react in
OCH,
I
389
390 R = AC 391 R = H
trifluoroacetic anhydride at 25". A by-product (390), the ester formed from 386 and 389, precipitated from the medium and mild alkaline hydrolysis of the soluble materials afforded 383. Alternative methods for the preparation of 383 consisted of either titanium tetrachloride (40-50% yield) or photoinduced (10-15 % yield) Fries rearrangement of 391. Improved methods for the oxidation of 383 to dehydrogriseofulvin (385) were also introduced. These involved carrying out the ferricyanide reaction with reverse addition or employing lead dioxide in acetone-ether solution. Both methods give 100% yields of product. Manganese dioxide in the same reaction gives 95-100% of 385, whereas silver oxide affords only $lo% of the dienone. The reduction of 385 to griseofulvin itself was carried out over a large amount of a 10% palladium-on-charcoal catalyst but yields (-36%) were not much superior to those obtained by Scott. A completely different approach to the synthesis of 361 was taken by Brossi and his groupies Scheme (40). They 0-alkylated the salicylic ester (392)
OCH3
OCH,
&lcH3 -
CH 3O
C
CI
392
+
H
3
0
c1
CO2CH3
4OCH ,CO,CH 3
393
OCH3 0
C0,CH3
CH 3O
397
+
R = CH.q
CI
-?+
394
399
400
Scheme 40
443
444
The Total Synthesis of Antibiotics
with methyl bromoacetate to give 393, then subjected 393 to a Dieckmann
condensation to obtain the coumaran-3-one 394. Michael addition of 394 to trans-3-penten-2-one in methanol using Triton B as a catalyst afforded 1, respectively, both two compounds, m.p. 184" and 164", in the ratio of 4: having structure 395. Cyclization of the dominant isomer by means of sodium methoxide in methanol (sodium ethoxide in ethanol did not work) led to the spiro compound, dl-epi-griseofulvin (396), having the reverse configuration at the spiro-center with respect to griseofulvin itself. Minor amounts of the by-products 397 and 398 were also isolated from the mother liquors of this reaction. Methylation of 396 with diazomethane afforded a mixture of 399 and 400 from which 399 was separated by chromatography. Partial conversion of 399 to dl-griseofulvin was accomplished by using the base-catalyzed equilibration originally discovered by M a ~ M i l l a n and I ~ ~ for which he proposed as one possibility the mechanism of Scheme 41. The
..
39 9
361
I
401
Scheme 41
isolation of small amounts of 401 suggests that this particular mechanism is correct. At equilibrium the system contains 60 % of dl-epigriseofulvin (399) and 40 % of dl-griseofulvin, which were separated by chromatography. dl-Griseofulvin was also synthesized by the base-catalyzed cyclization of the minor isomer of 395 followed by methylation of the product with diazomethane. However, the overall yield was extremely low. Resolution of dl-griseofulvin was carried out by using the brucine salt of griseofulvic acid followed by reconversion of the d-griseofulvic acid to 361 with diazomethane. The griseofulvic acid 388 was generated by the methanolic acid-catalyzed isomerization of dl-griseofulvin to dl-isogriseofulvin (402). which was then hydrolyzed by mild aqueous base. This procedure avoids the problem of
8. Miscellaneous Antibiotics
36 I
-
CH ,O &OW3 ('1
402
445
CH ,O CI
CH3
388
generating some epi-griseofulvic acid which would almost certainly be produced, via the equilibration of 361 and 399, if 361 were used directly in the base hydrolysis step. Finally, a very elegant synthesis of dl-griseofulvin has been published by Stork and Tomasz.16BThey elected to synthesize the spiro-ring system of 361 by means of a double Michael addition of a cross-conjugated vinyl ethynyl ketone to an active methylene compound. Based on this concept they felt that the specific use of the coumaranone 403 and of methoxyethynyl propenyl ketone (404) would directly produce the griseofulvin structure as folI0ws :
CH ,O
& 61
403
+
CH ,OC=CL
-0
CH,CH=CH
-*
404
CH30 An attractive feature of this proposition is that should it work, it would lead to the proper enol ether without the ambiguities of the diazomethane method used in the previous syntheses. The real difficulty in this approach was that compounds of type 404 were previously unknown largely because
446
The Total Synthesis of Antibiotics
of their great lability rather than to lack of a suitable preparative pathway. In fact the synthesis of 404 was solved in two steps. Condensation of the lithium salt of methoxyacetylene with crotonaldehyde at - 15' afforded the Li-C--C-OCH,
+ CH,CH=CHCHO
43-63%
OH
I
CH,CH=CHCH----CECOCH, 405
~
CH,OC=C
30%
>-0 CH~CH=CH 404
carbinol 405, which was then oxidized with activated manganese dioxide in methylene chloride. Provided the manganese dioxide were washed to neutrality prior to use consistent results were obtained. Preliminary condensation experiments with the ethoxy analog of 404 and diethyl malonate in the presence of potassium r-butoxide (catalytic amount) led to a substituted cyclohexenone identified as 406. Thus the initial concept was proven and 0
406
the condensation of the coumaranone 403 with 404 was now attempted under essentially the same reaction conditions. Chromatography of the product then afforded dl-griseofulvin (361) in 7 % yield. No dl-epi-griseofulvin (399) was detected in the reaction material, indicating that only the least stable isomer had been produced. The authors have suggested that the stereoselectivity observed here is the result of better overlap of the electron donor system in the transition state 407, which leads to dl-griseofulvin, than in the transition state 408, which leads to d/-epi-griseofulvin.
407
408
8. Miscellaneous Antibiotics
447
C. Novobiocin (362) Novobiocin is an antibiotic used largely against gram-positive organisms. Its structure was first defined completely by Shunk et al.,lee and V a t e r l a u ~ ~ ~ ~ reported its synthesis in 1964. Much of the difficulty in the synthesis of this molecule lay (a) in preparing a sugar moiety with the correct stereochemistry (Scheme 42) and (6) in determining proper conditions for the coupling of the protected sugar 419 to the coumarin nucleus. The synthesis of 419 began with 3,5,6-tri-0-benzyl-2-0-methyl-~-glucofuranose (409), which was oxidized with N-bromosuccinimide in the presence of barium carbonate to the lactone 410. Ring opening of 410 with methylamine afforded the amide 411, whose mesylate (412) when refluxed in aqueous acetic acid afforded 413, the 5-epimcr of 410. Treatment of 413 with methylmagnesium iodide gave the diol 414. Benzoylation of the latter esterified the secondary hydroxyl only, and the product 415 on catalytic
BzOqH,
BzOyH,
OCH3
OCH 409
410
BzOCH,
BzOCH2
N HCH 3
__f
OCHS
OCH
411
412
413
414 Scheme 42
448
The Total Synthesis of Antibiotics
I
I ___,
BzO
415
416
Scheme 42 (Continued)
reduction afforded the trio1 416. Lead tetraacetate cleavage of 416 followed by saponification of the benzoyl group produced noviose (417), the des-3-0-carbamoyl glycosidiccomponent of novobiocin. The methyl glycosides of 417 was treated by carbonyl chloride in pyridine to introduce the carbonate group. The mixture of glycosides (418) was then treated with acetyl chloride containing hydrogen chloride in nitromethane to give the key intermediate 2,3-U-carbonyl-noviosyl chloride 419. The required coumarin derivative 420 was synthesized by conventional methods (Scheme 43) and coupled successfully with 419 in quinoline in the presence of silver oxide and calcium sulfate to give the P-glycoside 421. The latter was reduced catalytically to remove the protective benzyl group on the C-4 oxygen atom and coupled with phenyldiazonium chloride with a view to introducing the 3-amino function on the coumarin ring. The product 422 was again reduced catalytically over palladium and the desired goal was reached. Acylation of the amine 423 with the acid halide 424 led to the penultimate compound 425 of the complete synthetic sequence. The final step, ammonolysis of 425, proceeded well, giving a mixture of novobiocin (362) and isonovobiocin (426) in which the former predominated. Fractional crystallization then gave pure novobiocin identical with the natural product.
FI
P
v, X
m d
6,/ I
v-4
N -3
449
+
450
8. Miscellaneous Anlibiotics
451
D. Cycloheximide (360)
Cycloheximide is a most interesting antibiotic from a biological point of view because it has such a broad spectrum of activity. In particular it has potent antifungal activity’?’ and is sold commercially for the control of such plant diseases as “dollar spot”, cherry leaf spot, and white pine blister rust. It is the most potent rodent repellent known, has antitumor activity, and is toxic to both protozoa and algae. Its activity springs from its ability to inhibit protein synthesis within the living cell. The structure of cycloheximide was determined by Kornfeld et a1.1?2and the stereochemistry by Johnson and his associates.173The latter group also carried out the only total synthesis reported114 so far for this molecule. In carrying out this synthesis stereoselectively, three major problems had to be solved: (a) the 4-methyl group of the cyclohexanone ring, which is remote from all other functionality, had to be built into the molecule in the unstable axial orientation; (b) the side-chain hydroxyl group had to be introduced in the (R)-configuration; and (c) the final steps of the synthesis had to be accomplished under fairly neutral conditions because of the sensitivity of cycloheximide to acid or base. A method of establishing the methyl groups in at least a trans relationship was suggested by the prior work of Williamson.175He had proposed that enamines of 2-substituted cyclohexanones had the 2-substituent in the quasi-axial orientation and offered this as the explanation for the difficulty that is observed when alkylation of such enamines is attempted. Any reagent approaching the double bond of the enamine would necessarily be involved in a nonbonded interaction with the quasi-axial substituent at the 2-position. The synthesis of cycloheximide required first a practical synthesis of glutarimide-3-acetic acid 429, and an improved method (Scheme 44) for
427
CH,CO,CH,
CH,O,C
h -
CH2C02H
CH2COCI
CN
428
H 429
Scheme 44
430
452
The Total Synthesis of Antibiotics
its preparation was reported by J o h n ~ o n ”in~ connection with a synthesis of actiphenol. Cope condensation of dimethyl acetonedicarboxylate with cyanoacetic acid led to the unsaturated nitrile, which when hydrogenated afforded 428. When the 428 was boiled with moderately concentrated hydrochloric acid followed by raising the reaction temperature to 235” the highly crystalline acid 429 was produced. Conversion of this to the acid chloride occurred smoothly in the presence of thionyl chloride containing a trace of dimethylformamide. In the critical step enamine formation with cis-2,4-dimethylcyclohexanone
434
433
435
0
R = AC R = COCHZCI
0
OCOCH2CI
OH
CH _.*
H
H
436
360 Scheme 45
8. Miscellaneous Antibiotics
453
led177to the enamine 431 of trans-2,4-dimethylcyclohexanone.Acylation of 431 with the acid chloride 430 (Scheme 45) gave, after hydrolysis with mild aqueous acid, dehydrocycloheximide (432).Reduction of 432 over a platinum catalyst very fortuitously occurred in such a way as to establish all five asymmetric centers of dihydrocycloheximide (433) stereoselectively (see below). An unexpected difficulty now arose. Although monoacetylation of 433 could be accomplished selectively at the side-chain hydroxyl group to give 434,whose oxidation by chromium trioxide gave cycloheximide acetate, no method could be found to hydrolyze the acetate group without destroying the molecule. If monoacylation of 433 were carried out with trifluoroacetic anhydride, only the compound with a trifluoroacetyl group on the ringhydroxyl could be isolated. Thus it appeared that if an unreactive monoester of 433 were prepared using a fairly weak acid, the ultimate product could not be hydrolyzed to give 360.If a strong acid anhydride were used, acylation of the side-chain hydroxyl undoubtedly occurred first, but was accompanied by ester migration to the ring-hydroxyl. The problem was solved by rapidly ' in the presence of exactly one acylating 433 with chloracetyl chloride at 4 equivalent of pyridine. Under these conditions ester migration was minimized and 435 was produced in good yield. Oxidation of 435 with chromium trioxide gave cycloheximide chloroacetate 436,which when hydrolyzed with potassium bicarbonate in aqueous methanol yielded cycloheximide (360). Both the dl- and I-forms of 360 were synthesized starting with the appropriate forms of cis-2,4-dimethylcyclohexanone. The reduction of 432 to 433 in such a specific manner requires some explanation. The authors consider that of the two possible conformers 432a and 432b of this diketone, the topside surface of 4328 presented less steric hindrance to catalyst approach than any of the other three. Reduction from this direction would then lead to 437 having the desired ring stereochemistry (Scheme 46). On the other hand, reduction from the least hindered side of 432a (i.e., from underneath) would generate 438. Preferential reduction of 437 could be expected to take place from the topside of 437 for steric reasons and assuming that the conformation of the side chain is controlled by hydrogen bonding between the side-chain carbonyl groups and the ring hydroxyl, then 439 would be (and is in fact) the expected product of the reaction. By similar reasoning, 438 would lead to 439. Some evidence for a two-step reduction was obtained by halting the reaction after the absorption of one equivalent of hydrogen, when 437 was isolated. Synthetic work in this area has also led to the synthesis of two isomers of cycloheximide, neocycloheximide~73and ol-epi-isocycloheximide,17*but their description is not warranted i n this essay.
1
CH3 HO
H CH3
P
H
H
RCO
P
H
H
H
1
1 H
CH 3 H
H
439 Scheme 46
454
H
438
OH
437
8. Miscellaneous Antibiotics
455
E. Mitomycin C (367) The mitomycins are a group of antibiotics that possess activity against both gram-positive and negative bacteria but are more noted for their antitumor properties. Only mitomycin C is used therapeutically. The structures of these compounds were deduced by Webb et a1.179and are unusual not only in the pyrrolo(1,Za)-indole ring that they contain but also because of the aziridine and aminobenzoquinone groups. Extensive synthetic work by a Lederle grouplaO led to the elaboration of 7-methoxymitosene (440), but efforts to prepare the tetracyclic nucleus of 367 were not successful. The aziridine ring is very unstable.
440
Recently a Japanese group has succeeded'8'"82 in preparing a compound having the complete ring system of the mitomycins. Their synthesis is shown in Scheme 47. Conventional methods were used to convert the indole 441 to
441
442
1----I
cH30m7&& 443
I
444
CO,CH
CH3
3
6 d 040%
'(0,I
445
Scheme 47
1
456
The Total Synthesis of Antibiotics
446
447 448
449
450 Scheme 47 (Continued)
the aldehyde 442, which on treatment with vinyltriphenylphosphonium bromide and sodium hydride in tetrahydrofuran led to the pyrroloindole 443. Carbomethoxylation with dimethyl carbonate and potassium 1-butoxide afforded 444, which was functionalized in the pyrroline ring with iodine azide preparatory to the introduction of the aziridine group. The iodo-azide 445 was reduced catalytically over palladium in methanol containing hydrogen chloride. The resulting amine hydrochloride 446 was cyclized by means of sodium methoxide in boiling methanol and yielded a crystalline mixture which was treated with methyl chlorocarbonate and triethylamine. From this reaction mixture the sought-after tetracyclic compound 447 was obtained by crystallization. There were also two by-products isolated to which structures 449 and 450 were assigned. The aziridino-pyrrolo( I ,2-a) indole 447 proved to be thermally unstable and rearranged above 150” to the oxazoline 448.
8. Miscellaneous Antibiotics
457
Further progress toward the synthesis of mitomycin C or related compounds has not been reported.
F. Chloramphenicol (358) Although chloromycetin or chloramphenicol is the last antibiotic to be discussed, it certainly is not the least important. Structurally it is one of the simplest antibiotics in everyday use, yet it has a very broad spectrum of activity. A serious side effect, however, is that it causes blood dyscrasias. Its structure determination183and initial synthesis184were carried out by the same group at Parke Davis. Several other syntheses have since been developed.186-189 The procedure discussed here, suitable for the large-scale production of 358, was devised by Ehrhart and his Condensation1eo of p-nitrobenzaldehyde with glycine methyl ester in methanolic hydrogen chloride gives exclusively the tltreo isomer of p-nitrophenylserine methyl ester (451) as its hydrochloride. Careful neutralization of this material with ammonia gave 451 itself, which was then acylated with dichloroacetyl chloride in aqueous sodium bicarbonate solution. The
451
DCH(OH)CHCOZCH. I
0
NO2
NHCOCHCI, 452
CH(OH)CHCON HN H,
I
NHCOCHCI, 453
358
63 %
458
The Total Synthesis of Antibiotics
amide ester 452 was converted to the hydrazide 453 by means of hydrazine hydrate in hot methanol. Treatment of 453 with nitrous acid led to the corresponding acyl azide, which without isolation was reduced with sodium borohydride in methanol to give dl-chloramphenicol 358. The biologically active form of 358 (D-form) was obtained in essentially the same way by using the L(+)-threo isomer of 451. Resolution of racemic 451 can be carried out using D(+)-tartaric acid. Chloramphenicol is the only antibiotic of therapeutic value that is more economically produced by total synthesis than by a fermentation process. From the point of view then of the commercial use of sophisticated chemistry to effect practical total syntheses, the record in the field of antibiotics is somewhat sparse if not sad, and the situation seems unlikely to change in the immediate future!
REFERENCES 1.
2. 3.
4.
5. 6. 7.
8. 9.
10. 11. 12. 13.
14. IS. 16.
(a) C. Y. Kao, in F. E. Russell and P. R. Saunders. Eds., Atiiirral Toxins. Pergamon Press, London, 1967, p. 109; (b) J. A. Vick, H. P. Ciuchta, and J. H. Manthei, p. 269. R. B. Woodward, in A. Todd, Ed., Perspectives irr Orgarric Clieiiiisfry. Interscience, New York, 1956,p. 156. Biochemically 6-aminopenicillanic acid can be obtained in the absence of side chain precursors [F. R. Batchelor, F. P. Doyle, J. H. C. Naylcr, and G. N. Rolinson, Nafroe, 183, 257 (1959)). It is also available from phenoxymethylpenicillin by treatment with kidney enzymes [US. Patent 3,070,511 (1962 to Lepetit)]. W. Cutting, Handbook 01Pharnracology. Meredith Corporation, New York, 1969, pp. 26-37. Summarized in The Cheniisfry of Penicillitr, Princeton University Press, Princeton, N.J.,1949. Reference 5, p. 1018. V. du Vigneaud, F. H. Carpenter, R. W. Holley, A. H. Livermore, and J. R. Rachaele, Science, 104,431 (1946). Reference 5, pp. 455-472. Reference 5, pp. 806-807. Reference 5, p. 909. J. C. Sheehan and K. R. Henery-Logan, J . A m . C h i t . Soc., 79, 1262 (1957); 81, 3089 (1959). J. C. Sheehan and D. A . Johnson, J . h i . Clietti. Soc., 76, 158 (1954). J. C.Sheehan and K.R. Henery-Logan, J. A m . Chetn. SOC.,81, 5838 (1959). A. J. Base, G. Spiegelman and M. S. Manhas, J . Am. Chern. Soc., 90,4506 (1968). 1. Ugi, Airgew. Chem. INI.Ed. Etrgl., 1, 8 (1962). E. J. Corey and A. M. Felix, J . Am. Chern. Soc., 87, 2518 (1965).
References 17.
18. 19. 20. 21. 22. 23. 24.
25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47.
459
R. B. Woodward, K. Heusler, J. Gosteli, P. Naegeli, W. Oppolzer, R. Ramage, S. Ranganathan, and H. Vorbruggen, J. Am. Chem. SOC.,88,852 (1966). R. B. Morin, B. G.Jackson, R. A. Meuller, E. R. Lavingnino, W. B. Scanlon, and S. L. Andrews, J. Am. Chent. Soc., 91, 1401 (1969); 85, 1896 (1963). J. A. Webber, E. M. van Heyningen and R. T. Vasileff, J. Am. Cheni. Soc., 91, 5675 (1969). D. 0.Spry, J. Ani. Chem. SOC., 92, 5007 (1970). G. E. Woodward and E. F. Schroeder, J. Am. Chem. SOC.,59, 1690 (1937). R. Heymbs, G. Arniard, and G. Nomin6, Contpr. rend. C , 263, 170 (1966). C. Mannich and M. Bauroth, Ber., 57, 1108 (1924). R. R. Chauvette, E. H. Flynn, B. G. Jackson, E. R. Lavagnino, R. B. Morin, R. A. Mueller, R. P. Pioch, R. W. Roeske, C. W. Ryan, J. L. Spencer, and E. van Heyningen. J. Am. Cfieni.SOC.,84, 3401 (1962). J. R. D. McCormick, E. R. Jensen, P. A. Miller, and A. P. Doerschak, J. Am. Chem. SOC..82, 3381 (1960). L. H. Conover, K. Butler, J. D. Johnston, J. J. Korst, and R. B. Woodward, J. Am. Chem. Soc., 84, 3222 (1962); 90,439 (1968). R. B. Woodward, Pure Appl. Chem., 6, 561 (1963). H. Muxfeldt and W. Rogalski, J. Am. Chem. SOC.,87, 933 (1965). H. Muxfeldt, E. Jacobs, and K. Uhlig, Cheni. Ber., 95,2901 (1962). H. Muxfeldt, G. Buhr, and R.Bangert, Angew. Clieni. Itlt. Ed. En& 1, 157 (1962) A. I. Gurevich, M. G. Karapetyan, M. N. Kolosov, V. G. Korobko, V. V. Onoprienko, S. A. Popravko, and M. M. Shemyakin, TefruhedrotiLeu., 1967, 131. H. H. Inhoffen, H. Muxfeldt, H. Schaefer, and H. Kramer, Croaf. Chem. Acfa., 29, 329 (1957). M. N. Kolosov, S. A. Popravko, and M. M. Shemyakin, Ann., 668,86 (1963). A. 1. Gurevich, M. G . Karapetyan, and M . N. Kolosov, Khitn. Prirodn. Sedin., Akud. Nuuk. LIZ. S.S.R., 2, 141 (1966); Chenr. Abs., 65, 13627 (1966). M. Schach von Wittenau, J. Org. Cheni., 29, 2746 (1964). A. I. Scott and C. T. Bedford, J. Am. Chem. SOC.,84, 2271 (1962). H. Muxfeldt, G. Hardtmann. F. Kathawala, E. Vedejs, and J. B. Mooberry, J . Am. Chenr. SOC.,90, 6534 (1968). H. Muxfeldt. J. Behling, G. Grethe, and W. Rogalski, J. Am. Chem. SOC., 89, 4991 (1967). H. Muxfeldt, Atrgew. Cheni., 74, 825 (1962). Other examples of this type of condensation are given in reference 38. A. S. Kende, T. L. Fields, J. H. Boothe, and S. Kushner, J. Am. Chetii. Soc., 83, 439 (1961). D. H. R. Barton, Chenr. Brit., 6,301 (1970). See also Barton and coworkers, J. Chem. SOC.(C) 1971, 2164-2241. A. Schatz. E. Bugie, and S. A. Waksman, Proc. SOC.Exp. Biol. Med., 55,66 (1944). M. L. Wolfrom, S. M. O h , and W. J. Polglase, J. Am. Cfiem. SOC., 72, 1724 (1950). F. A. Kuehl, Jr., E. H.Flynn, F. W. Holly, R. Mozingo, and K. Folkers, J. Atri. Cfieni.Soc., 69, 3032 (1947). J. R. Dyer, W. E. McGonigal, and K. C. Ricc, J. Ani. Chem. Soc., 87, 654 (1965). H. Maehr and C. P. Schaffner, J. Am. Chem. Soc., 92, 1697 (1970).
460
The Total Synthesis of Antibiotics
48. H. Hoeksema, A. D. Argoudelis, and P. F. Wiley, J . Am. Client. Soc., 84, 3212 (1962). 49. M. Nakajima, A. Hasegawa, and N. Kurihara, Tefraliedron Leff.,1964,967. 50. M. Nakajima, I. Iornida, and S. Takei, Cheni. Ber., 92, 163 (1959); M. Nakajima, A. Hasegawa and N. Kurihara, Chern. Ber., 95, 2708 (1962). 51. M. Nakajima, A. Hasegawa, N. Kurihara, H. Shibata, T. Ueno, and D. Nishimura, TefrahedronL e f t . , 1968,623. 52. S. Umezawa, K. Tatsuta, and S . Koto, BUN.Cheni. Sac. Japan, 42, 533 (1969). 53. S. Koto, T.Tsumura, Y . Koto, and S . Urnezawa, Bull. Clietn. Soc. Japan, 41,2765 (1968). 54. H.H.Baer, Chon. Ber., 93, 2865 (1960);J . Ant. Cheni. Soe., 83, 1882 (1961). 55. S. Koto, K. Tatsuta, E. Ketazawa, and S. Umezawa, BUN. Cheni. SOC.Japan, 41, 2769 (1968). 56. Compound 156 was prepared from 6-acetarnido-6-deoxy-~-glucose[F. Cramer, 0. Ollerbach and H. Springman, Chetn. Ber., 92,384 (1959)l in a manner completely analogous to that used for the conversion of 148 to 143. 57. S. Umezawa and S . Koto, Bull. Chem. Soe. Japan, 39, 2014 (1966). 58. P. F.Lloyd and M. Stacey, Chem. Ind., 1955, 917. 59. S. Umezawa, S. Koto, K. Tatsuta, H. Hineno, Y. Nishimura, and T. Tsumura, Bull. Clieni. Soc. Japan, 42,537. 60. S. Umezawa, K. Tatsuta, and T. Tsurnura, Bull. Cheni. SOC.Japan, 42, 529 (1969). 61. T.Suhara. F.Sasaki, K. Maeda, H. Urnezawa. and M. Ohno, J. Ani. Cliern. SOC., 90,6559 (1968). 62. M. Nakajima, H. Shibota, K. Kitahara, S. Takahashi, and A. Hasegawa, Tetrahedron Lett., 1968,2271.
63. Y. Suhara, K. Maeda, H. Umezawa, and M . Ohno, TefrahedronL e f t . , 1966, 1239. 64. E. Vis and P. Karrer, Helu. Chini. Acfa 37, 378 (1954). 65. R. R.Herr, H. K. Jahnke, and A. D. Argoudelis,J. Am. Cheni. SOC.,89,4808(1967). 66. E. Hardegger. A. Meier, and A. Stoos, Helu. Chiin. Ada, 52, 2555 (1969). 66a. B. J. Magerlein, R. D. Birkenrneyer, R. R. Herr, and F. Kagan, J. Ani. Chern. Sac., 89, 2459 (1967). 66b. H. Hoeksema, B. Bannister, R. D. Birkenrneyer, F. Kagan, B. J. Magerlein, F. A. MacKellar, W. Schroeder, C. Slornp, and R. R. Herr, J. Ani. Chem. SOC.,86,4223 (1964); R. R. Herr and G. Slomp, J. h i . Cheni. SOC.,89, 2444 (1967). 66c. M. Bethell, G.W. Kenner, and R. C. Sheppard, Nafrtre, 194,864 (1962). 66d. B. J. Magerlein, Tetrahedron L e f f . ,1970 3 3 . 66e. G.B. Howarth, W. A. Szarek, and J. K. N. Jones, J . Chern. Sac. ( C ) , 1970, 2218 66f. D. G . Lance, W. A. Szarek, J. K. N. Jones, and G . B. Howarth, Can. J. Chetri., 47, 2871 (1969). 66g. G. B. Howarth, D. G . Lance, W. A. Szarek, and J. K. N. Jones, Cart. J. Chetn., 47, 75 (1969). 66h.H.Saeki and E. Ohki, Cheni. P h r t n . Bull. (Tokyo) 18, 789 (1970). 67. G. 0.Gorton, J. E. Lancaster, G. E. van Lear, W. Fulmor, and W. E. Meyer, J. Am. Chem SOC.,91, 1535 (1969);D.A. Shurnan, R. K. Robins, and M. J. Robins, J. Am. Chetn, SOC.,91, 3391 (1969).
References
461
68. K. G. Cunninghan, S. A. Hutchison, W. Manson, and F. S. Spring, J. Chem. Soc., 1951, 2299. 69. J. D. Dutcher. M. H. von Saltza, and F. E. Pansy, Antiniic. Agfs. Chemother. 1963, 83; H.Agahigian, G. D. Vickers, M. H. von Saltza, J. Reid, A. I. Cohen, and H. Gauthier, J. Org. Chenr. 30, 1085 (1965). 70. K. Isono, K. Asahi, and S. Suzuki, J. Am. Cheni. Soc., 91,7490 (1969). 71. T. Naka, T. Hashizume, and M. Nishimura, Tetrahedron Lett., 1971, 95. 71a. J. J. Fox, Y . Kuwada, T. Ueda, and E. B. Whipple, Antiniic. Agts. Chemother., 1964,518;J. J. Fox, Y.Kuwada, and K. A. Watanabe, TetrahedronLert., 1968,6029; K. A. Watanabe, M. P. Kotick, and J. J. Fox, Chem. Pharm. Bull. (Tokyo), 17, 416 (1969). 72. J. W. Hinman, E. L. Caron, and C. DeBoer, J. Am. Chem. Soc., 75, 5864 (1953); S. Hanessian, and T. H. Haskell, Tetrahedron Lett., 1964,2451. 73. H.Taniyama and F. Muyoshi, Chem. Pharm. Bidl. (Tokyo), 10, 156 (1962); E. E. van Tamelen, J. R. Dyer, H. A. Whaley, H. E. Carter, and G. B. Whitfield, Jr., J. Ani. Cheni. Soc., 83, 4295 (1961); H.E. Carter, C. C. Sweeley, E. E. Daniels, J. E. McNary, C. P. Schnaffner, C. A. West, E. E. van Tarnelen, J. R. Dyer, and H. H. Whaley, J. Am. Chem. Soc., 83,4296 (1961). 74. S.Takemura, Chem. Pharm. Bull. (Tokyo), 8, 574, 578 (1960). 75. B. R. Baker, R. E. Schaub, J. P. Joseph, and J. H.Williams, J. Am. Chem. Soc., 77, 12 (1955). 76. B. R. Baker, R. E. Schaub, and J. H. Williams, J. Am. Chew Soc., 77, 7 (1955). 77. B. R. Baker, J. P. Joseph, and J. H. Williams, J. Am. Chem. Soc., 77, 1 (1955). 78. W. Schroeder and H. Hoeksema, J . Am. Clrem. Soc., 81, 1767 (1959). 79. M. L.Wolfrom, A. Thompson, and E. F. Evans, J. Am. Chem. Soc., 67,1793 (1945). 80. J. FarkaS and F. S&m, Tetrahedron Lett., 1962, 813. 81. J. R. McCarthy, Jr., R. K. Robins, and M. J. Robins, J. Am. Chem. Soc., 90,4993 (1 968). 82. E. Walton, F. W. Holly, G . E. Boxer, R. F. Nutt, and S. R. Jenkins, J, Med. Chenr., 8, 659 (1965). 83. C.D. Anderson, L. Goodman, and B. R. Baker, J. Am. Chem. Soc., 80,5247 (1958). . Soc., 91, 2102 84. R. L. Tolman. R. K. Robins, and L. B. Townsend, J. A I ~ ICheni. (1 969). 85. R. L.Tolman, R. K. Robins, and L. B. Townsend, J . Heterocyc. Cheni., 4, 230 (1967). 86. T.Hashizurne and H . Iwamura, Tetrahedron Lett., 1965,3095. 87. K. Ohkuma, J. Antibiotics (Tokyo), 13A, 361 (1960). 88. 0.Theander, Actu Chem. Scand., 18, 2209 (1964). 89. K. A. Watanabe, R. S. Goody, and J. J. Fox, Tetrahedron,26, 3883 (1970); M. P. Kotick, R. S. Klein, K. A. Watanabe, and J. J. Fox, Carbohyd. Res., 11,369 (1969). 90. K. A. Watanabe, M. P. Kotick, and J. J. Fox, J. Org. Chem., 35, 231 (1970). 90a. K. A. Watanabe, I. Wempen, and J. J. Fox, Chem. Phami. Bull. (Tokyo), 18,2368 (1970). 90b. E.J. Reist, R. R. Spencer, D. F. Calkins, B. R. Baker, and L. Goodman, J. Org. Cheni., 30, 2312 (1965).
462 91. 92. 93.
The Total Synthesis of Antibiotics H. Yonehara and N. Otake, Antittricrob. AXIS. Chemother., 1965, 855. P. Baudet and E. Cherbuliez, Helu. Chitn. Acta, 47, 661 (1964). Y. Hirolsu, T. Shiba, and T. Kaneko, Bull. Chetti. SOC.Japum, 43, 1870 (1970), and preceding papers.
94. D. L. Swallow and E. P. Abraham, Biochem. J., 72, 326 (1959); W. Stoffel and L. C. Craig, J . Am. Chetti. Sor., 83, 145 (1961). 95. H. Vanderhaegl and G. Parmentier. J. Anr. Chetn. SOC..82, 4414 (1960). 96. G. R. Delpierre, F. W. Eastwood, G. E. Gram, D. G. I. Kingston, P. S . Sarin, L. Todd, and D. H. Williams, J. Chetn. Soc. (C), 1966, 1653. 97. B. W. Bycroft, D. Cameron, A. Hassanali-Walji, and A. W. Johnson, Tetrahedron Lett., 1969, 2539. 98. E. Schroder and K. Liibke, Experientia, 19, 57 (1963). 99. P. Quitt, R. 0. Studer, and K.Vogler, Helu. Chini. A r m , 47, 166 (1964). 100. D. W. Russell, J. Ciiem. SOC.,1962, 753. 101. H. H. Wasserman. J. J . Keggi, and J. E. McKeon, J. Am. Chem. SOC.,83, 4107 (1961); 84, 2978 (1962). 102. R. L. Hamill, C. E. Higgens, N. E. Boaz, and M. Gorman, Tetrahedron Lett., 1969, 4255. 103. H. Brockmann and H. Geeren, Anti., 603, 216 (1957). 104. M. M. Shemyakin, N. A. Aldanova, E. I. Vinogradova, and M. Yu Feigina, Tetrahedron Lett.. 1963, 1921. 105. E. Schroder and K. Lubke, The Peptides, Vol. 11. Academic Press, New York, 1966, pp. 396 ff. 106. R. D. Hotchkiss, Ado. Enzyniol., 4, 153 (1944); R. L. M. Synge, Q. Reu. (London), 3 , 245 (1949). 107. R. D. Hotchkiss and R. J. Dubos, J . Biol. Chem., 132, 79 (1940). 108. J. D. Gregory and L. C. Craig, J. Biol. Chern., 172, 839 (1948). 109. S. ishii and B. Witkop, J. Am. Chem. Soc., 85, 1832 (1963); see also L. K. Ramachandran, Biochern., 2, 1138 (1963). 110. R. Sarges and B. Witkop, J. Am. Chetti. Soc., 87, 2015 (1965). 111. R. Sarges and B. Witkop, J. Am. Chem. Soc., 87, 2020 (1965). 112. R. Sarges and B. Witkop, J. Am. Chem. Soc., 87,2027 (1965). 113. G. F. Gause and M. G. Brazhnikova, Lancet If, 1944, 715. 114. R. L. M.Synge, Biochem. J., 39, 363 (1945); R. Consden, A. H. Gordon, A. J. P. Martin, and R. L. M. Synge, Biochenr. J., 40, xliii (1946); 41, 596 (1947). 115. R. Schwyzer and P. Sieber, Helu. Chini. Acta, 40, 624 (1957); Angew. Chem., 68, 518 (1958). 116. K. Kurahashi, J . Biochem. (Tokyo), 56, 101 (1964); S. Otani and Y. Saito, J. Biochem. (Tokyo), 56, 103 (1964). 117 A. R. Battersby and L. C. Craig, J. Anr. Chetti. Soc., 74, 4023 (1952); A. Paladini and L. C. Craig, J. Am. Chenr. Soc., 76, 688 (1954). 118. T. P. King and L. C. Craig, J. Am. Chettr. Soc., 77, 6627 (1955). 119. M. A. Ruttenberg, T. P. King, and L. C. Craig, Biochetti., 4, 11 (1965). 120. M. Ohno, T. Koto, S. Makisumi, and N. Izumiya, BUN. Chem. SOC.Jupan, 39, 1738 (1966).
References
463
Japan, 43,2199 (1970). 121. K. Kuromizu and N. Izumiya, Bull. Client. SOC. Japan, 43,2944 (1970). 122. K. Kuromizu and N. Izumiya, Bull. Chem. SOC. 123. N. Mitsuyasu, S. Matsuura. M. Waki, M. Ohno, S . Makisumi, and N. Izumiya, Bull. Chem. SOC. Japan, 43, 1829 (1970). 124. K. Fujikawa, T. Sakamoto, T. Suzuki. and K. Kurahashi, Biochim. Riopltyys. Acta, 169, 520 (1968). 125. R. Schwyzer and P. Sieber, Helu. Chim. Acta, 41,2186 (1958). 126. For a summary of the structure work see reference 105, pp. 451-472. A review of the chemistry of these compounds has also appeared. K. Vogler and R. 0. Studer, Experientia, 22, 345 (1966); S. Wilkinson and L.A. Lowe, Nature, 188, 311 (1966). 127. S. A. Waksman and H. B. Woodruff, J. Bacteriol., 42,231 (1941). 128. H. Brockmann, Angea. Chem., 66, 1 (1954): 72, 939 (1960); Fortsckr. Chem. Org. Naturst., 18, 1 (1960); Naturwiss., 50, 689 (1963). 129. H. Brockmann and H. Lackner, Naturwiss., 51, 384 (1964); Chem. Ber., 101, 1312 (1968). 130. H. Brockmann, H. Lackner, R. Mecke, G. Troemel, and H. S . Petras, Chem. Ber., 99,717 (1966). 92, 3771 (1970). 131. J. Meienhofer, J. Am. Chem. SOC., 132. Yu. A. Ovchinnikov, V. T.Ivanov, and I. 1. Mikhaleva, Tetrahedron Lett., 1971, 159. 132a. M. M. Shemyakin. Yu A. Ovchinnikov, V. K. Antonov, A. A. Kiryushkin, V. T. Ivanov, V. I. Shchelokov, and A. M. Shkrob, Tetrahedron Lett., 1964, 47; A. A. Kiryushkin, V. I. Shchelokov, V. K. Antonov, Y. A. Ovchinnikov, and M. M. Shemyakin, Khim. prirod. Soedinenii, 3, 267 (1967). 133. P. F. Wiley, K. Gerzon, E. H. Flynn, M. V. Sigal, Jr., 0. Weaver, U. C. Quarck, R. R. Chauvette, and R. Monahan, J. Am. Chem. SOC., 79,6062 (1957). 134. P. W. K. Woo, H. W. Dion, and Q.R. Bartz, J . Am. Chent. Sor., 86, 2726 (1964). 1964, 842. 135. M. L. Dhar, V. Thaller, and M. C. Whiting, J. Cltenr. SOC., 136. A. C. Cope, U. Axen, E. P. Burrows, and J. Weinlich, J. Am. Chem. Sor., 88,4228 (1966). 137. A. J. Birch, C. W. Holzapfel, R. W. Rickards, C. Djerassi, M. Suzuki, J. Westley, J. D. Dutcher, and R. Thomas, Tetrahedron Lett., 1964, 1485. 138. R. N. Mirrington, E. Ritchie, C. W. Shoppee, W. C. Taylor, and S . Sternhell, Tetrahedron Lett., 1964,365; F. McCapra, A. I. Scott, P. Delmotte, and J. DelmottePlaqute, Tetrahedron Lett., 1964, 869. (C), 1967, 1913. 139. P. M. Baker, B. W. Dycroft, and J. C. Roberts, J. Chew. SOC. 140. R. Huisgen and M. Reitz, Tetrahedron, 2, 271 (1958). 141, D. Taub, N. N. Girotra, R. D. Hoffsommer, C. H. Kuo, H. L. Slates, S. Weber, and N. L. Wendler. Tetrahedron, 24,2443 (1968). 142. 1. Vlattas, 1. T. Harrison, L. Tokes, J. H. Fried, and A. D. Cross, J. Org. Cheni., 33,4176 (1968). 143. C. H. Hassell, Org. Reactions, 9 (1957). 144. 1. J. Borowitz, G. Gonis, R. Kelsey, R. Rapp, and G. J. Williams, J. Org. Chem., 31, 3032 (1966). 145. J. F. Bagli and H. Immer, Can. J . Chem., 46, 3115 (1968).
,
464
The Total Synthesis of Antibiotics
146. C. H. Stammer, A. N. Wilson, F. W. Holly, and K. Folkers, J . Ant. Chem. Soc., 77, 2346 (1955);PI. A. Plattner, A. Boller, H. Frick, A. Fiirst, B. Hegedus, H. Kirchensteiner, St. Majnoni, R. Schlapfer, and H.Spiegelberg, Helu. C h i . Actu, 40, 1531 (1957). 147. H. Kawaguchi, H. Tsukiura, M. Okanishi, T. Ohmori, K. Fujisawa, and H. Koshiyama, J. An/ibio/ics, A18.1 (1965);H. Kawaguchi, T . Naito, and H. Tsukiura, J. An/ibio/ics, A18, 11 (1965). 148. F.Arcamone, G. Franceschi, P. Orezzi, G. Cassinelli, W. Barbieri, and R. Mondelli, J. A m . Chem. Soc., 86, 5334 (1964); F. Arcamone. G. Cassinelli, P. Orezzi, G. Franceschi, and R. Mondelli, J. Ani. Chem. SOC.,86, 5335 (1964); F. Arcamone, G. Franceschi, P. Orezzi, and S. Penco, Tefrahedroti L e f f . ,1968,3349;F. Arcamone, G. Cassinelli, G. Francexhi, P. Orezzi, and R. Mondelli, Te/rahedrort Left., 1968, 3353. 149. I). S. Tarbell, R. M. Carman, D. D. Chapman, K. R. Huffman, and N. J. McCorkindale, J. Am. Chem. Soc.. 82, 1005 (1960); N. J. McCorkindale and J. G. Sime, Proc. Chetti. Soc., 1961, 331. 150. W. 0.Godtfredtsen and S. Vangedal, Tefruhedron, 18, 1029 (1962);D. Arigoni, W. von Daehne, W. 0. Godtfredsen, A. Melera, and S. Vangedal, Experienfiu, 20, 344 (1964); D.Arigoni, W.yon Daehne, W. 0. Godtfredsen, A. Marquet, and A. Melera, Experienrin, 19, 521 (1963). 151. A. Agtarap, J. W. Chamberlin, M. Pinkerton, and L. Steinrauf, J . Am. Chein. Soc., 89, 5737 (1967). 152. W. Oppolzer, V. Prelog and P. Sensi, Experieniiu, 20, 336 (1964). 153. K.V. Rao, K. Biemann, and R. B. Woodward, J. Am. Clrem. Soc., 85,2532 (1963). 154. R. R. Herr, J. Anr. Chem. Soc., 81,2595 (1959);H. E. Hennis, L. G. Duquette, and F.Johnson, J. Org. C/iern., 33, 904 (1968). 155. G.Biichi and J. E. Powell, Jr., J. Ani. Chern. SOC., 92, 3126 (1970). 156. Unpublished work by the author. 157. W. Leimgruber, A. D. Batcho, and R. C. Czajkowski, J . A m Chern. Soc., 90, 5641 (1968). 158. W. Leimgruber, V. StefanoviC, F. Schenker, A. Karr, and J. Berger, J. Ant. Chem. Soc., 87, 5791 (1965). 159. W . Leimgruber, A. D. Batcho, and F. Schenker, J. Am. Chem. Soc., 87, 5793 (1965). 160. M. Artico, G. De Martino, G. Filacchioni, and R. Giuliano. Farm. Ed. Sci., 24, 276 (1969). 161. E. Oxford, H. Raistrick, and P. Simonart, Biochewt. J . , 33, 240 (1939). 162. J. F. Grove, J. MacMillan, T. P. C. Mulholland, and M. A. T. Rogers, J . C/ietti. Soc.. 1952, 3977. 163. A. C. Day, J. Nabney, and A. 1. Scott, J . C/ietn. Soc., 1961, 4067. 164. D. H. R. Barton and J. Cohen, in FesfsrhriJr A . Sfoll, Birkhauser, Bade, 1957,p. 117. 165. D. Taub, C. H. Kuo, H. L. Slates. and N. L. Wendler, Terrahedron, 19, 1 (1963). 166. A. Brossi, M. Baumann, M. Gerecke, and E. Kyburz, Helu. Chirn. A d a , 43, 2071 (1960). 167. J. MacMillan, J. Chenr. Soc., 1959, 1823. 168. G. Stork and M. Tomasz, J. Atti. Client. Soc., 86, 471 (1964).
References
465
169. C. H. Shunk, C. H. Stammer, E. A. Kaczka, E. Walton, C. F. Spencer, A. N. Wilson, J. W. Richter, F. W. Holly, and K. Folkers, J. Am. C/tem. Soc., 78, 1770 (1956). 170. B. P. Vaterlaus, J. Kiss, and H. Spiegelberg, Helu. Chin?. Acra, 47, 381 (1964); B. P. Vaterlaus, K. Doebel, J. Kiss, A. 1. Rachlin, and H.Spiegelberg, Helu. Cltim. A m , 47, 390 (1964); B. P. Vaterlaus and H. Spiegelberg, Helu. Chiin. A d a , 47, 508 (1964). 171. A. J. Whiffen, N. Bohonas, and R. L. Emerson, J. Bact., 52,610 (1946). 172. E. C. Kornfeld, R. G. Jones, and T. V. Parke, J. Am. Client. SOC.,71, 150 (1949). 173. F. Johnson, N. A. Starkovsky, and W. D. Gurowitz, J. Am. Chem. Soc.. 87, 3492 (1965); F. Johnson, N. A. Starkovsky, and A. A. Carlson, 87,4612 (1965). 174. F. Johnson, N. A. Starkovsky, A. C. Paton, and A. A. Carlson, J. Ant. Client. Soc., 88, 149 (1966). 175. W. R. N. Williamson, Tefrahedrott,3, 314 (1958). 176. F. Johnson, J. Org. Chem., 27, 3658 (1962). 177. F. Johnson and A. Whitehead, Tetrahedron Left., 1964, 3825; H. J. Schaeffer and V. K. Jain, J. Org. Cliem., 29, 2595 (1964). 178. F. Johnson, A. A. Carlson, and N. A. Starkovsky, J. Org. Chem., 31, 1327 (1966). 179. J. S. Webb. D. B. Cosulich, J. H.Mowat, J. B. Patrick, R. W’. Broschard, W. E.
Meyer. R. P. Williams, C. F. Wolf, W. Fulmor, C. Pidacks, and J. E. Lancaster.
180.
181. 182. 183. 184. 185. 186. 187. 188. 189.
190.
J. Am. Chem. Soc., 84, 3185, 3187 (1962). G. R. Allen, Jr., J. F. Poletto, and M. J. Weiss, J . Org. Chern., 30, 2897 (1965); G. R. Allen, Jr., and M. J. Weiss, J . Org. Chem., 30, 2905 (1965); W. A. Remers, R. H. Roth and M. J. Weiss, J. Org. Chem., 30, 2910 (1965). T. Hirata, Y.Yamada, and M. Matsui, TefrahedronLeff.,1969, 19. T.Hirata, Y.Yamada, and M. Matsui, TefrahedrortLeft.,1969,4107. M. C. Rebstock, H.M. Crooks, Jr., T. Controulis, and Q. R. Bartz, J . Am. Chem. Soc., 71, 2458 (1949). J. Controulis, M. C. Rebstock, and H. M. Crookes, Jr., J. Am. Cliem. SOC.,71, 2463 (1949). L. M. Long and H. D. Troutman, J . Am. Chem. Soc., 71,2469 (1949). L. M. Long and H. D. Troutrnan, J. Am. Chem. Soc., 71,2473 (1949). G.Ehrhart, W. Siedel and H. Nahm, Cltem. Ber., 90,2088 (1957). S. Urnezawa and T. Suami, Bull. Chem. SOC.Japan, 27,477 (1954). V. A. Mikhalev, M. I. Dorokhova, N. E. Smolina, A. M. Zhelokhovtseva, A. P. Skoldinov, A. I. Ivanov, A. P. Arendaruk, M. I. Galchenko, V. A. Skorodumov, and D. D. Smolin, Anfibiofiki, 4, 21 (1959). E. D. Bergmann, H. Bendas, and W. Taub, J. Client. SOC.,1951,2673.
Total Synthesis OfNatural Products, Volume1 Edited by John Apsimon Copyright © 1973 by John Wiley & Sons, Inc.
The Total Synthesis of Naturally Occurring Oxygen Ring Compounds F. M. DEAN University of Liverpool, England
1. Introduction 2. Chromenes 3 . Sterigmatocystin and the Aflatoxins 4. The Rotenoids
5. Coumestans and Pterocarpans
6 . Amphipyrones (Pyronoquinones)
7. Xanthones
References
467
468 485 498 513 525 534 555
1. INTRODUCTION
The types of compound discussed in this chapter form but a small proportion of those available, and the reader may wish to know the grounds on which the 467
468
The Synthesis of Oxygen Ring Compounds
selections were made. Compounds or methods already multiply reviewed (e.g., flavones, oxidative cyclization) were rejected in favor of less welldocumented areas of research provided these were currently lively. Then it was decided to discuss several syntheses within a few groups of compounds (rather than single syntheses from each of numerous groups) in order to maintain coherence and illustrate the subtler variations in problems and in methods of attack. Usually, the selected compounds contain more than one oxygen ring and all are derived from phenolic nuclei in order to bring out the fact that, not infrequently, problems in heterocyclic chemistry are quite as much problems in phenol chemistry, the heterocyclic part being the easier. And finally, of course, the choice was personal. This chapter contains about 100 syntheses of compounds falling into 6 main classes : chromenes, dihydrofurobenzofurans, chromanochromanones, benzofurochromans, xanthones and pyronoquinones. It is proposed to rename this last group the amphipyrones. 2.
CHROMENES
We are concerned here with compounds containing the nucleus of 2,2dimethyl-2H-chromene 1, of which a large number occur in the higher plants.'e2 To produce this nucleus the biosynthetical pathways that, separately, produce terpenes or phenols must merge,3 and one result is a very great range of complexity. In gambogic acid4 2, for example, a phenolic kernel carries three C, residues and one C , , residue, each being modified in a different fashion thus presenting a considerable synthetical problem.
1
2 Gambogic acid
Good methods for synthesising chromenes are all recent discoveries. A synthesis of lapachenoleK 3 shows the use of an older m e t h ~ d , ~still -~ sometimes used, in which coumarins were treated with a Grignard reagent and the resulting alcohol cyclized:
2. Chromenes
469
3 Lapachenole
The final cyclization has to be effected without the catalytic aid of acids. An ever-present difficulty in working with chromenes is their sensitivity to acids, especially protic acids, which, presumably, may lead to carbonium ions of types 4a and 4b,both of which are stabilized by interaction with the oxygen atom:
4b
4a
Ions of type 4a must underlie the cyclization of gambogic acid 1 to gamboginic acid 5 and could explain the isolation from the essential oil of Ageratum conyzoides not only of ageratochromene 6 but also of a dimerl0 with structure 7. On the other hand the complex polycyclic s t r u c t ~ r e s ~ ~ ~ ~ ~ obtained by treating lapachenole 3 and other chromenes with acids probably require the intervention of ions of type 4b.
Gambogic acid
6 Ageratochromene
5 Gamboginic acid
7
The most widely used chromene synthesis was that introduced by Spath" in which the requisite phenol was heated with 2-methylbut-3-yn-2-01 either alone or with some very mild Lewis acid catalyst such as zinc chloride. The
470
The Synthesis of Oxygen Ring Compounds
method succeeded with seselin, xanthyletin, luvungetin, and some others, but it is enough to say that 2% was considered to be an acceptable yield. Eflorts to use C,, acetylenes were even less rewarding,I2 and not all phenolic nuclei withstood the treatment. In a synthesis of pongachrornenelg 8 attempts to add the pyran ring to the flavone 9 failed entirely, whereas the more roundabout method gave a yield of 4 . 1 % :
0
8 Pongachromene
9
AllanRohinwn
HO’
+
OH
OMe
OMc 0
0
0 (rcjcctcd)
0 10 Lonchocarpine
11 Seselin
In a closely similar preparation of lonchocarpine 10, NickP secured the catalytic effect of the cation in a medium of very low acidity by using zinc carbonate instead of zinc chloride, while other workerP have greatly improved the preparation of seselin 11 from umbelliferone (7-hydroxycoumarin) by omitting a catalyst altogether but using 1 ,Zdichlorobenzene
2. Chromenes
471
as the solvent at 160'. But even so the yield was only 6%, and the technique is now wholly replaced by a modern version. In this, the acetylenic alcohol is converted into 3-chloro-3-methylbutyne which is used to etherify the phenol in the presence of potassium carbonate with iodide catalyst, and the ether is heated in diethylaniline for some hours. Discovered by Iwai and Ide,16 the method affords chromenes directly in yields of 80% or more. The illustration is for evodionol methyl ether 12 obtained thus by Hlubucek, Ritchie, and Taylor,'' who prepared 7-methoxy2,2-dimethylchromene, ageratochromene 6, luvungetin, and seselin 11 in the same way.
12 Evodionol methyl ether
The rearrangement is very interesting since, if viewed as an allylic rearrangement, neither the parent acetylene nor the allenic transition state (or intermediate) appears to have a suitable geometry. Nevertheless, an allenic intermediate is indeed responsible and can be trapped internally by an electrocyclic reaction leading to a tricycle-octane derivative18 13:
p-
Me
1
13
The resemblance of 13 to the bridged part of gambogic acid 2 inspires speculations to be taken up in Section 7, but it is the electrocyclic ring closure of 14 that we have to notice particularly here, since it turns out to be a feature of nearly every chromene synthesis. A useful corollary is provided by the partial reduction of the acetylenic ethers to allylic ethers. These upon conventional Claisen rearrangement affordo-dimethylallylphenols*in excellent
* It is frequently convenient to refer to the 3-methylbut-2-enyl group as 3,3-dimethylallyl, or as dimethylallyl only, or even simply as prenyl.
472
The Synthesis of Oxygen Ring Compounds
overall yields. Osthenol 15, demethylsuberosin, coumurrayin, and other compounds have been prepared thus,'@ and the method is generally much superior to most direct alkylation techniques :
Osthenol
15
Such 0-3,3-dirnethylallylphenolsare very common, and it has long been re~ognizedl-~ that related dimethylallylphenols and 2,2-dimethylchromenes occur in the same plant or in closely allied species and believed that one structural type may be a biosynthetical precursor of the other. Some attention has been paid to the reduction of chromenes to allylphenols, a change that can be effected in simple cases by the Birch reduction,20or photochemically,a1 but much more has been devoted to the reverse process of oxidative cyclization, an idea first suggested by Ollis and Sutherland' and later modified by Turnerz2and others. Whatever the details, the central theme is the production of a quinone methide that cyclizes spontaneously: -2H ----+
\
Such a cyclization was first achieved in uitro during studies on mycophenolic the reagent being dichlorodicyanobenzoquinone (DDQ) in refluxing benzene. This system is known2*to be able to operate by hydride abstraction from suitable hydrocarbon groupings, which suggests a sequence such as the following: H H+
Col
[OH]-
2. Chromenes
473
This question of mechanism is important, for it seems that cyclizations to chromenes would be on the same initial footing as other dehydrogenations which, consequently, might well interfere. Though little is known about the matter, the danger can be assessed from an example of dehydrogenation taken from the cannabinoid series:25
This reaction takes but 2 hours for 90% completion, whereas most chromene cyclizations require something between 1 and 16 hours. Cyclizations by DDQ have been evaluated mainly by Italian workerszB and usually give yields around 50%. Where the starting dimethylallylphenol is itself a readily available natural product such yields are not unsatisfactory, and the method is at its best in these circumstances. When the dimethylallylphenol has to be synthesized, the approach seems much less attractive, since the C-alkylation of phenols is often a confused affair not giving more than about 20% of the desired product. A common technique to introduce the dimethylallyl group is to employ the necessary phenol as its lithium salt, which is heated in benzene with 4-bromo-Zmethylbut-2-ene. These conditions are designed to minimise 0-alkylation, but there is a tendency for them to induce a concomitant cyclization to a chroman which, as explained later, may not be a viable intermediate. This criticism applies even more strongly to “biogenetically modeled” alkylations employing phosphates instead of bromides,27these often giving chromans in yields as high as 80 %:
Examples2s of the successful use of the route include the preparations of alloevodionol methyl ether 16 and of (f)-cannabichromene 17, the latter exemplifying the insertion of a C,, side chain, and both exemplifying the directive effects of hydrogen bonded carbonyl groups :
The Synthesis of Oxygen Ring Compounds
474
Me
\
Me
c- o.,
M e O + j J o ?
C,H,,Br ___+
McO
Me0
Alloevodionol methyl ether 16
110
’
17 Cannabichromene
...Q 1
\
HO’
\
I
\
OH
It must be noted, however, that ideas are changing rapidly in this area. It is now recognized that the cyclization of o-prenylphenols may be a relatively slow process, and that keeping the temperature as near to room temperature as possible and the reaction time as short as possible may be all that is necessary to obtain good yields of the desired alkenylphenols.zeClearly, the acidity of the medium is not as crucial as was thought originally, though the reaction is certainly acid catalyzed. Hence the success of the technique introduced by Bohlmann and Kleineas who report that treating phenols with 3-methylbut2-enyl alcohol and boron trifluoride etherate in dioxan at about 25” gives
2. Chromenes
475
good yields of their C-prenyl derivatives. The method is used increa~ingly,9~.~~ as in a first synthesiss0 of evodionol (from Evodia littoralis and Melicope
simplex) :
OMc
""0"" Preremirol
MeCO \
OMc
1 pM: Evodionol
A D Q
OMc
Acronylin
If the dimethylallylation of the phenol leads inevitably to a chroman the DDQ reaction may still be applicable if there is a free hydroxyl group in a position allowing quinone-methide formation.sa The point can be made with a synthesis of alloevodionol 18. The unusual pyranochromene franklinone 19, however, seems exceptional, for it is reported to be formed by DDQ oxidation of the corresponding tetrahydro-derivative with no free hydroxyl groups.28 Previously, franklinone had evaded synthesis altogether. One last point of interest is that chelation will protect a phenolic hydroxy group to some extent and may therefore direct a DDQ cyclization as required for s y n t h e ~ e s ~of~the v ~ trimethyl ~ ether 20 of flemingin C:
18 Alloevodionol
19 Franklinone
20 Flemingin C trimethyl ether
476
The Synthesis of Oxygen Ring Compounds
If the DDQ dehydrogenation fails to convert the chroman into a chromene it may still be possible to fall back on bromination-dehydrobromination techniques; one such instance is discussed near the end of Section 7. Another form of oxidative cyclization is known in which a quinone nucleus can be regarded as an internal oxidant. It is catalyzed by bases and occurs with ease, and was discovered, largely accidentally, during a "purification" of ubiquinone on alumina columns that actually gave ubichromenol 21 instead :34835
Me Ubiquioone
0
1
21 Ubichromenol
The mechanism outlined again emphasizes spontaneous cyclization in an oquinone methide. Beside alumina, potassium hydroxide has been used as the catalyst and, much better, pyridine at its boiling point, this technique inducing a 92% c o n v e r s i ~ n .Sodium ~~ hydride in benzene has also been ~ecomrnended.~'Perhaps more significantly for plant life processes, the same change can be induced by irradiation with visible light,se a minor consequence of which is that TLC of such quinones must be conducted in the dark. If UV light is used, the quinone is mainly reduced to the quinol instead. We turn now to a different type of chromene synthesis, one based on the fact that chromanones are rather easily accessible. Ageratochromene 22 provides o u r illustration of the appr0ach,3~(as shown on top of page 477). Meerwein-Ponndorf reduction may replace the hydride reduction. The last step, dehydration, is, however, critical, since the acid catalysts usually utilized prejudice the survival of the One solution has already been noted. Another is to use anhydrous copper 11 sulfate,40but the problem has been better solved as in the case of acronycine 23 by the use of phosphorus oxychloride in refluxing ~ y r i d i n e . ~In* the case of xanthoxyletin 24 it was even possible to use sodium hydrogen sulfate provided the product was sublimed away from the reaction area.41This synthesis is illustrated on account of an interesting anomalous carbonyl reduction by borohydride, of clausenin
23 Acronycine
OH
22 Ageratochrornene
25 yielding the hydrocarbon as well as the expected alcohol. Presumably
borohydride solutions are alkaline enough to open the lactone ring so that the way is open to quinone methide formation and reduction; upon work-up, the lactone ring would reclose spontaneously as is usual with coumarins. Of course, alcohol dehydration problems can be side-stepped altogether by pyrolyzing the acetates, as in a synthesis of lapachenole 3 via the related ~hromanone.~
momo J
25 Clauscnin
NaBH,
I)CHINl
\
\
OMe
/
24 Xnntlioxyletin
2) NsHSO,
\
/
HO HO
1
NaBH,
478
The Synthesis of Oxygen Ring Compounds
3-Hydroxychromans are available naturally, as reduction products of chroman-3-ones or, most recently, as thallic oxidation products of ally1 aryl ethers:"" but acid-catalyzed dehydration is liable to induce skeletal rearrangements and prototropic shifts leading to isomers of the desired c h r o r n e n e ~ . ~ ~ ~~ and j a t m a n ~ i n o l . ~ ~ Examples occur in the chemistry of ~ e l i n e t i n(lomatiol) The example shown concerns d e ~ u r s i n '26, ~ a coumarin constituent of an old Chinese medicinal herb; protic acids provoke a rearrangement leading to anhydronodakenetin 27 while the tosylate in collidine affords the chromene, xanthyletin 28 (as shown on opposite page).
26
Decursin
27 A nhydronodakenetin
J 28 Xanthyletin
Unlike most chromens, siccanochromene A 30 is a fungal product. It comes from Helminthosporium siccans, a plant pathogen. An interesting synthesis of this compound by Nozoe and HiraiqBcontains steps of importance for the present discussion, and may be summarized as opposite. The lithium derivative was used to attain an orientation not easily available by direct substitution reactions, but it is the spontaneous dehydration of intermediate 29 that invites attention since the contrast with the conditions for dehydrating 4-hydroxychromans is so marked. Again the formation and cyclization of a quinone methide allows a satisfactory rationalization for chromene ring formation.
I
\
(2 stereoisomers)
2 29
CH, =PPhs
\
OH
\
\
HO
30 Siccanochromene
479
480
The Synthesis of Oxygen Ring Compounds
If flindersine 31 be admitted to the honorary degree of chromene, then one synthesis4’ of it supplies another instructive variation upon the same theme:
0
31
Flindersine
The two preceding reactions serve to introduce one of the most powerful of all chromene syntheses. Essentially, the phenol (as phenoxide ion) reacts with an unsaturated aldehyde giving just that type of alcohol which, like 29, spontaneously generates a quinone methide and therefore the chromene ring system.
2.
Chronienes
481
The reaction is induced by pyridine at 100-150" for several hours. It can be regarded as an aldol condensation occurring in the ketonic form of the phenol, and i n fact it succeeds only with those phenols (resorcinol, phloroglucinol, and their derivatives) that are well known for this kind of tautomerism. Monohydric phenols and quinol are not amenable. Announced by Crombie and P ~ n s f o r d , ~one ~ "variation of the method was early used to synthesize certain constituents of hashish, the drug extracted from Cannabissatioa. Condensed with citral, olivetol yields several products,dsh of which only the simplest, cannabichromene 32, is produced in fair yield. This fact and the orientation can be attributed to the difficulty of inserting one large group ortho to another:
C,H,,
32 Cannabichromcnc
Such reactions may go further, however, for some of these chromenes are sensitive to bases and the final results often depend upon the conditions used, especially the amount of pyridine. Although some details are not yet ~ e t t l e d , ~ ~two * ~ cyclizations O are formulated speculatively to emphasize yet again the central role of quinone methide chemistry as shown overleaf:
A
C'ai1n;tbichromene
482
Cil rylidcnc-cannahin
2. Chromenes
483
A stepwise sequence is indicated for the cyclization to the cyclobutane derivative, cannabicyclol, by the steric factor which makes an electrocyclic reaction under Woodward-Hoffman rules very doubtful. A photochemical cyclization would be possible, of course, and is known.61 The alkaloid mahanimbine 33 offers a nitrogen analog of the situation, and in fact this alkaloid can be synthesized and transformed similarly:61#6a
33 Mnhanirnbinc
Such further cyclizations and other misadventures can be rendered unimportant by masking the responsible hydroxy group by chelation. Thus the followingcondensation affords in 67 % yield a chromene 34 suitable for elaboration into ethers of the flemingins A, B, and C:
0
0 35
34
484
The Synlhesis of Oxygen Ring Compounds
Even the trienal, farnesal, undergoes a similar reaction4**giving 42% of the desired chromene 35. In a further refinement, unsaturation is avoided and the aldehyde replaced by an acetal function. Rather higher reaction temperatures seem necessary, but resins are much reduced and yields improved. Very satisfactory syntheses of evodionol methyl ether 12, lonchocarpin 10, and many other chromenes have been achieved Illustrated are reactions leading to jacareubin 36 and to acronycine 37, the latter, obtained from Acronychia baueri, being of particular interest for its antitumor activity. Further examples would include
+
b'covo PY
Mco
0
0
36 Jacareubin
0
*+&pyJ H
('
=
\
H
37 Acronycine (R = Me)
some of the pyranocoumarins3* that are found in various species of Maminea. Another recent papers4 describes additional cases of dual annelation as well as raising the old specter of dimerization in a new as well as the old forms-a claim that cyclization occurs at a chelated hydroxy group is perhaps not justified, however.
3. Sterigmatocystin and the Aflatoxins
485
Of other methods of synthesizing chromenes, that of greatest interest centers on an ylide reaction. It has one advantage over the methods just oulined in that it begins with a salicylaldehyde derivative and therefore leaves no doubt as to the orientation of the product. The sodium salt of the derivatiOe itself acts as the base initiating a Wittig reaction in dimethylformamide as solvent:
The primary product is believed to be the open-chain system shown, since hydrogenation at the early stage leads to the corresponding o-alkylphenol. The cyclization occurs at a later stage but exactly how is not clear. Yields of 40 % are ~ b t a i n a b l e . ~The ~ " alkaloid girinimbine 38, which occurs with mahanimbine in Murruyu koenigin, has been synthesized in this fashion and its constitution defined thereby:65b
3. STERIGMATOCYSTIN AND THE AFLATOXINS
A small but important group of fungal metabolites contain a furo[2,3-b]benzofuran nucleus, In addition to this unusual structural feature, the high toxicity of all and especially the marked carcinogenic powers of some of these compounds has excited much interest in them.66.57Sterigmatocystin6*(39), from some strains of Aspergillus oersicolor (Vuill.) Tiraboschi, was the first
486
The Synthesis of Oxygen Ring Compounds
member of the group to be recognized. The versicolorins are related anthraq u i n o n e ~ and , ~ ~ so is dothistromin,eOa plant pathogen implicated in pine needle blight. The notorious carcinogens, the aflatoxins,el are produced by Aspergillus fravus and A . parasiticus: two have been synthesized, B, with structure 40 and B, with structure 41. Another aflatoxin, aflatoxin M I42, comes from the milk of cows or the urine of sheep or the liver of rats that have ingested aflatoxin B, and has also been ~ynthesized.~'Some strains of A. fraous produce sterigmatocystin derivatives instead of, or as well as, aflatoxins, thus demonstrating a very close biosynthetical relationship between the two series. Other close relationships are believed to exist with the anthraquinone aversin, the methyl ether of which has very recently been synthesized as described at the end of this section.
*y
39 Sterigmatocystin
0
\
/
OMe
41 Aflatoxin B,
&
40 Aflatoxin B,
0
\
OMe /
42 Aflatoxin M,
Some direct methods of creating a furobenzofuran are known. Examples include (a) treatment of a suitable chloroethyl lactone with an amine:Oa
q
3. Sterlgmatocystin and the Aflatoxins
(b) condensation of glyoxal with a phenol :63
2
~0 + '~ (CH0l2
+
R
and ( c ) condensation of dihydrofuran with acetylquinone :a4
1
487
R
Ac
But since no direct method seemed sufficiently adaptable, relatively long sequences have been used in practice, often with 5,7-dihydroxy-4-methylcoumarin 43 as the starting point.* Methylationes of 43 with dimethyl sulfate and base affords mainly the 7-methyl ether 44. This selectivity depends on the fact that the ion 45 is para quinonoid and therefore more stable than the alternative, which is ortho quinonoid. Hence, provided that only one equivalent of base is used, the 7oxide is chiefly present at the site of reaction. In agreement, the UV spectrum of the dihydroxycoumarin is nearly the same as that of the 5-methyl ether but different from that of the 7-methyl ether when all are examined in limited base.e6 By the same argument, on the other hand, the electrons of the 7hydroxy group in the unionized dihydroxycoumarin are less readily available
43 R = H
45
44 R = M e
46
The synthesis of the 3-(2-hydroxyphenyl)furan system and a cyclization by electrolytic oxidation have been achieved but details are not yet available.64a
488
The Synthesis of Oxygen Ring Compounds
than those of the 5-hydroxy group because of the para-quinonoid interaction indicated in 46, so benzylation by sources of benzyl carbonium ion should occur elsewhere preferentially. This seems to explain the formatione6 of the 5-benzyl ether when the potassium carbonate-acetone technique is used. Either way, the mixed ether 47 is accessible, and is o x i d i ~ e dby ~ ~selenium .~~ dioxide6*to the yellow aldehyde 48 without damage to the benzyl methylene group. Acetal49 is produced by the orthoformate method and then controlled hydrogenation saturates the olefinic bond taking the nucleus into the dihydrocoumarin series as 50. At the same time the benzyl group is lost yet the acetal ether links survive, quite the reverse selectivity to the selenium dioxide reaction. Lithal reduction followed by acid hydrolysis now yields the tetrahydrofurobenzofuran derivative 51 as the racemate, but otherwise identical with (optically active) degradation product of ~terigmatocystin.~~ The ring fusion is necessarily cis because of the considerable strain in the trans fused arrangement. Furthermore, Knight, Roberts, Roffey, and SheppardsS have condensed the furobenzofuran derivative 51 with ethyl cyclopentanone-2carboxylate by Pechrnann's method but with hydrogen chloride in ethanol instead of the usual sulfuric acid. (In phloroglucinol derivatives the electronreleasing effects of the three hydroxy groups reinforce one another so that the nucleus becomes very highly nucleophilic and only the mildest catalysts are needed to induce electrophilic substitutions, the more powerful ones merely inducing further, unwanted reactions.) The product 52 is the racemate corresponding to a reduction product of aflatoxin B, 40. The same workers70 have also condensed the phenol 51 with ethyl 3oxoadipate, thus obtaining a coumarinylpropionic ester and thence, by alkaline hydrolysis, the acid 53. Alkaline hydrolysis leaves the acetal function untouched, of course, but hazards the lactonic system since this can open to a cis-cinnamic acid in which geometrical isomerization is known to occur with loss of viable material. In fact, losses were not serious, and the synthesis was continued by using oxalyl chloride to secure the acid chloride and aluminum chloride at -5" to cyclize it to (f)-aflatoxin B, 54 in 33 % yield. Such cyclizations are general.61*70-72 During the final stages great care was taken to ensure the survival of the acetal grouping under acidic attack. Acyclic acetals would probably not survive at all, the entropy factor being against it, but with cyclic ones such as these, ring opening does not, in itself, present a major difficulty since ring closure is always the preferred sequel. More acute problems are posed by the synthesis of the unsaturated acetals, sterigrnatocystin 39, and aflatoxin B, 40,in which acid-catalyzed additions to the vinylic ether group are a constant danger. This can be avoided by introducing the unsaturation at the last step, but even here conditions have
PhCH,O
48
PhCH,O
CH-0
CH(OEt),
49
"""% H
EtO
OEt
th
4 \e O m o CH(OEt),
HO 50
489
490
The Synthesis of Oxygen Ring Compounds
ca
COt Et
EtoYo
0 HCI OMe 52
51
two steps
HO 53 been found that make it possible to operate elsewhere in a molecule containing an exposed vinylic acetal group. The synthesis of (&)-ailatoxin B, was announced by Buchi and his colleaguesss in 1966. Their starting point was the aldehyde 48 alluded to earlier. Coumarins are rather easily reduced by zinc and acid, and in this one the double bond is activated further by the aldehyde carbonyl group so the reduction readily yielded the dihydrocoumarin aldehyde 55. The next step was taken inadvertently, however, for dihydrocoumarins are rather easily hydrolyzed, here spontaneously exposing the latent aldehyde function thus allowing hemiacetalS6 formation and relactonization : 54
M
c
o
PhCHtO 55
q CHO
o
M
e
0
PhCHtO
-
'0 9
OH
/o OH
e- M
"'0%
OH 57
0
e
PhCHzO
O
q
0
3. Sterlgmatocystin and the Aflatoxins
491
Such "/I-acyl lactone" rearrangements are well known, having been extensively studied by K ~ r t e . ' ~ Catalytic debenzylation produced the phenolic lactone 57, and attempts were made to elaborate this into a coumarin using ethyl methyl 3-oxoadipate in a conventional Pechmann condensation under the influence of 86% sulfuric acid. This overpowerful reagent largely catalyzed ring-opening to a benzofuran 58, thus exposing the system to a host of undesirable consequences, many of which took their toll. In a gentler procedure, the lactone 57 gave with methanolic hydrogen chloride the acetal-ester 59, which was not isolated but allowed to react at once with ethyl methyl 3-oxoadipate thus smoothly producing a coumarin 60 in yields approaching 60% (see page 492). Next, hydrolysis gave the lactonic acid 61, which, with oxalyl chloride and then aluminum chloride, afforded the cyclopentenocoumarin 62. The crucial last stages were now at hand. Selective reduction of the y-lactone carbonyl group giving the hemiacetal63 and dehydration of this to give (f)-aflatoxin B, 40 were needed. Neither stage has proved to be particularly easy. Disiamylborane does attack the y-lactonic function in preference to the others (the infrared band at 1790 cm-I may perhaps indicate an electrophilicity higher than is usual in most lactones) and the hemiacetal63 can be obtained in about 20% yield. Since dehydration cannot be accomplished directly, the acetate has to be pyrolyzed at 240" for I5 minutes, a process yielding 40 % of (f)-aflatoxin B, 40. The racemate has not yet been resolved. Identities have been established by racemizing natural aflatoxin B, hydrate, which is easily obtained by treating the natural product with trifluoracetic acid and water. In base, the acetal-hemiacetal system opens liberating the aldehyde function; consequently, theadjacent optically active center* becomes labile and thecompound racemises :
Subsequently, Buchi and Wein~eb'~ evolved a most convincing synthesis of aflatoxin M, (42), provision for the angular hydroxy group being made a t the very beginning by commencing with the coumaranone 64. The methylene group was brominated using phenyltrimethylammonium perbromide in tetrahydrofuran, and the halogen replaced using benzyl alcohol in the presence of calcium carbonate (seemingly, this is an S,1 reaction in which the
0
4
0 c)
E
h
:
0
492
4
0
0
r;i
S
3. Sterlgmatocystin and the Aflatoxins
493
carbonium center can be developed next to a carbonyl group, no doubt because of the considerable stabilizing effect of the ether oxygen atom). Treatment with allylmagnesium bromide and oxidation with the sodium periodate-osmium tetroxide reagent then supplied the aldehyde 65 as a mixture of stereoisomers :
64
At this point the benzyl group had to be detached without disturbing either the aldehyde group or the angular hydroxy group, which is itself benzylic. This delicate manoeuvre was accomplished catalytically with the help of palladium-charcoal in ethyl acetate. Cyclization was spontaneous, as expected, and acetylation furnished the diacetate 66,pyrolysis of which in a short-contact, continuous-flow system at 450" gave the vinylic ether 67 in 75 % yield.
66
68a R = OH 68b R = CI
67
69
70
494
The Synthesis of Oxygen Ring Compounds
Phloroglucinol esters being susceptible to hydrolysis, the phenol was obtained from the acetate 67 and sodium hydrogen carbonate and attachment of the coumarin system investigated. At first sight all that is required is a Pechmann condensation with the cyclopentadienone ester 68a, but this ester is well known to be inert in this respect.75 We might speculate that protonation would yield ion 69 upon which hydrogen bonding and charge distribution as in 70 would confer considerable stability. However, the derived chloride 68b readily condenses when zinc carbonate is used both as catalyst and acid scavenger and produces (f)-aflatoxin M, 42, identical as nearly as may be with the natural product.
Aflatoxin G2
73a
-
74
M e*
6
R =H 73b R =Me
72
71
E O t:0 75
yo2 76
OMe
Another member of the aflatoxin series, aflatoxin G2,has structure 71. Bycroft, Hatton, and RoberW have commenced explorations of routes to structures of this kind and succeeded in preparing lactone 72 as a model for the novel section of the G2molecule. Direct condensation between a salicylaldehyde derivative 73a and, for example, ethyl acetoacetate, is a longestablished method of preparing substituted coumarins, but is variable when
3. Sterigmatocystin and the Aflatoxins
495
applied to acetophenones such as 73b and generally useless when applied to higher homologs. In the present case, the difficulty was cleverly overcome by arranging for entropy factors to counterbalance it. Phenol 74 was esterified by ethyl chlorocarbonylacetate and the product 75 readily cyclized to the coumarin 76 in the presence of piperidine. Cold concentrated sulfuric acid effected a second cyclization supplying the desired model 72. Unfortunately, corresponding reactions were not observable upon transfer to the phloroglucinol or even to the resorcinol series. In parallel studies, Rance and Roberts?' have provided syntheses of (&)0-methylsterigmatocystin 77 and of its dihydro derivative 78, which was later found to be a natural product in its own right in an optically active form.?* The only problems not already considered are those implicit in securing a xanthone nucleus. Phenol 51 condensed with methyl 2-bromo-6methoxybenzoate giving the diphenyl ether 79 when the catalyst was cuprous chloride in refluxing pyridine. When the corresponding acid was treated with oxalyl chloride in dichloromethane it cyclized at once to the desired xanthone, ( f)-dihydro-0-methylsterigmatocystin 78, which is a tribute to the nucleophilicity of the phloroglucinol nucleus. For the synthesis of (&)-0methylsterigmatocystin 77, a xanthone ring was attached similarly to the acetal-ester 59, the first stage giving 80. Alkaline hydrolysis and acid treatment led to the lactone 81 converted by the oxalyl chloride method into the xanthone 82. Then selective reduction by disiamylborane, acetylation, and pyrolysis served to complete the task.
0
OMe
0
( f)-0-Methylsterigmatocystin
Dihydro-0-methylsterigmatocystin
79
80
77
78
496
The Synthesis of Oxygen Ring Compounds
0
0
/
\
0 81
OMe
82
The anthraquinone, aversin 83a, is of particular interest because anthraquinones of this type are likely to be the biogenetic precursors of the aflatoxins and sterigmato~ystins.~~ It has been obtainedso from the mold Aspergillus versicolor (Vuillemin) Tiraboschi, which also produces the versicolorins and divers related compounds, and its methyl ether 83b has recently been synthesisedeoby a route that takes advantage of one of the newest techniques for building up hydroxyanthraquinone derivatives.8L Commonly, anthraquinones are prepared by Friedel-Crafts condensation of an aromatic substrate with a derivative of phthalic anhydride, the first stage being the formation of a benzoylbenzoic acid. This is often easily accomplished, but the next step, the cyclization forming the quinone ring, now demands electrophilic attack on an electron-deficient system and is often impossible without conditions of a stringency that precludes the method here. The new technique utilizes basic media,*I and the present example starts at phenol 51 which, with oxalyl chloride in methylene chloride, furnishes the lactone 84, treatment with methanol then giving the ester 85, and methylation followed by hydrolysis the glyoxylic acid 86. Such acids are not of great stability and in many of their reactions lose carbon monoxide. Thus thionyl chloride produces the acid chloride 87, and a Friedel-Crafts acylation with 3,5-dimethoxybenzyl cyanide as substrate is now possible, the product being the benzophenone derivative 88. The cyclization to the cyanoanthracene is effected with methoxidc ion in dimethylformamide or in dimethylsulfoxide. It is presumed that a carbanion is formed and extrudes a methoxy group by an addition-elimination mechanism made possible partly by the presence of the benzophenone carbonyl group and partly by the fact that the aromaticity of phloroglucinol derivatives renders them but little more stable than conjugated but nonaromatic analogs would be. Either methoxy group may be extruded, so the cyanoanthracene 89 is accompanied by the isomer with angular annelation. The isomers are readily distinguished by spectroscopic means, and the requisite
83a R = H;Avtrsin 83b R E Me
84
OMc
CI
I10 87
‘ 0
Me0
86
+I
\
014
0
85
OMe
88
89
0 90
497
498
The Synthesis of Oxygen Ring Compounds
one oxidized by alkaline hydrogen peroxide to aversin methyl ether 83b. Probably the last reaction involves hydroxylation-elimination sequences as suggested in diagram 90. 4. THE ROTENOIDS
With structure 91, rotenone is at once the most important and nearly the
most complex member of this group of plant products, munduserone 92 being the simplest, and all being characterized by what has usually been called a chromanochromanone nucleus. * The rotenoids are physiologically active, stunning fish and killing aphids and other plant pests, and are the active constituents of derris dust. Several synthetical routes are now available, although to date no fully satisfactory synthesis of rotenone itself has been described and no attempt has been made to synthesise amorphin 93, the only glycoside yet discovered in this series.82 The foundations of rotenoid synthesisaawere laid before 1940 by numerous contributions, often overlapping, from several distinguished groups of chemists; and we are forced to take much of this work for granted in order to concentrate on the more recent results. Throughout the earlier studies, 8'
91 Rotenone
92
Munduserone
* For rotenone, Chemiicul Ahrrucrs gives the name 1,2,I2,12u-tetrahydro-2-isopropenyl-
8,9-dimethoxy(l]benzopyrano[3,4-b]furo[2,3-h][I]benzopyran-6(6uH)-one.Another system of nomenclature is based upon the parent structure r o t o ~ e n : * ~
4. The Rotenoids
499
93 Amorphin
ideas were dominated by the fact that gentle oxidation (iodine and sodium acetate is the favorite reagent, but manganese dioxide and ferricyanide are also used) of the chromanochromanone nucleus dehydrogenates it to the chromenochromone nucleus :
q=q& \
\
The importance of this is that the new nucleus is readily dismantled by alkaline hydrolysis giving products that are relatively accessible synthetically :
In reverse, such a sequence constitutes a general route to synthetic rotenoids. It has been usual to start with a suitably substituted salicylaldehyde and convert it via the azlactone synthesis into a benzyl cyanide:
500
The Synthesis of Oxygen Ring Compounds
0
0
The next step, a particularly critical one, is the Hoeschcondensation between the cyanide and a phenol which, since it has to be derived from resorcinol, phloroglucinol, or pyrogallol, is highly nucleophilic and reacts under the influence of hydrogen chloride with only mild catalysts such as zinc chloride or, sometimes, with no catalyst at all. The resulting imine is hydrolyzed by warm aqueous acid to the desired ketonic acid:
CO2Et I
C02Et
I
HO'
OH
R
Although this approach to such ketonic acids is in some respects clumsy, it is fair to say that it is nearly always successful. An alternative, in which the phenylacetic acid itself is condensed with the phenol by means of polyphosphoric acid, is simpler and has been used as shown as part of a synthesis of munduserone,85 though it is ambiguous and failed in another case?
OMe
4. The Rotenoids
501
The dual cyclization necessary to convert the ketone into the chromenochromone is effected in one step by heating with acetic anhydride and sodium acetate: if the phenoxyacid group is esterified, the cyclization fails so it is likely that a mixed anhydride is an essential intermediate:86s87
In the absence of labile substituents the yields may be as high as 50%; and the method has sufficed for the synthesissBof racemic deguelin, for example, via dehydrodeguelin 94. Occasionally the ester is cyclized instead of the acid, but the reagent is then sodium ethoxide and the reaction constitutes, in part, a Dieckmann condensationBSgiving the aroylchroman-3-one 95, which cyclizes easily. The problem of reducing the chromenochromone nucleus to the chromanochromanone nucleus had barred progress. Though hydrogenation can be successfulse~QO it is not easily controlled and, in any case, nearly all rotenoids contain other reducible double bonds that would be attacked. Matsui and Miyanonl solved this problem with sodium borohydride, which reduces not only the carbonyl group but also the olefinic bond leaving a chromanol which is commonly oxidized to the required ketone by the Oppenauer method or, sometimes, by manganese dioxide used under conditions minimizing the further dehydrogenation leading back to the chromenochromone, (as shown on top of next page).
502
The Synthesis of Oxygen Ring Compounds
This reduction of an olefinic bond by borohydride ion may be understood in terms of a 1,Caddition which is greatly promoted because it leads to an enolic stilbene system with very extensive conjugation. Simple carbonyl reduction would result only in much decreased conjugation. After ketonization of the enol the phoenix carbonyl group can be reduced in a second step:
HX
X-
In theory, the ring fusion in the chromanochromanone nucleus can be cis or trans. However, the natural compounds are all cis fused and it is cis fusion only that is found in the synthetic materials. Models s h o that ~ in~ the trans arrangement there is an unavoidable collision between the carbonyl oxygen atom and the hydrogen atom at position 1 on the neighboring aromatic ring. In the more flexible cis arrangement there need be no such collision. Since there is probably little difference otherwise in energy between the cis and trans arrangements, this factor is enough to account for the somewhat unusual result.
~
~
~
4. The Rotenoids
503
To synthesize rotenone itself by this route, the half nitrile 96 of derric half ester has to be condensed with tubanol 978 by means of hydrogen chloride, but attempts to do this directly fail because tubanol is sensitive to acids and can add hydrogen chloride, suffer scission of the allylic ether system, and isomerize into the more aromatic and more stable (but still sensitive) furan 97b. The difficulty was minimized by the use, instead of tubanol, of the alcohol 98 prepared by the sequence shown from the benzofuran 99. The Hoesch condensation then furnished a mixture of products from which the derrisic ester analog 100 was isolated in less than 8 % yield and dehydrated with phosphorus tribromide in cold pyridine, this step giving the methyl ester 101 of (f)-derrisic acid in about 10% yield. The acid itself was needed for resolution, and the rather stringent conditions (potassium hydroxide in refluxing ethanol) reduced the yield yet further, only 10 mg of half-crystalline acid being obtained from 70 mg of crude ester. In these
97a
96
97b
99
98
OH /
\
0
100
tHCl
96
+ 98
--., 'OMe OMe
OMe 101
504
The Synthesis of Oxygen Ring Compounds
circumstances resolution could not be attempted, and the ingenious course was adopted of emecting an inverse resolution. Active derrisic acid, obtained by degradation of rotenone 91, was observed to resolve (f)-1-phenylethylamine, one salt separating from ethanol. Consequently, the resolution of the synthetic derrisic acid by active phenylethylamine must also be possible, though it was not actually demonstrated. The final stages of rotenone synthesis had all been demonstrated earlier using materials from degradation; that is, cyclization of (optically active) derrisic acid to the chromenochromone (dehydrorotenone), borohydride reduction, and Oppenauer oxidation. This synthesis of rotenone is hardly satisfying, and in a sense it is formal rather than actual. At the moment, however, it is the onh synthesis, other routes not having been fully worked out. Moreover, the route has been greatly improved recentlys3 by the use of dicyclohexylcarbodiimide (DCCD) to effect cyclization of derrisic acid 102 to dehydrorotenone 103. No doubt this cyclization also involvcs a mixed anhydridees but it does not require acidic conditions so although a number of by-products complicate the issue, the propenyl grouping survives and the yields are good.
Derrisic acid
Dehydrorotenone
\ 102
103
DCCD
6Me 104
4.
The Rotenoids
505
The postulated intermediate aroylchroman-3-one 104 above suggests a way of assembling the rotenoid skeleton so as to avoid entirely the Hoesch condensation and its attendant difficulties. Heated with sodium acetate and acetic anhydride, derric acid 105 readily cyclizes to the requisite chromanone 106, which at once forms the enol acetateB4107. Herbert, Ollis, and RussellB5used piperidine to regenerate the chromanone from the acetate and simultaneously catalyze its condensation with 2-hydroxy-Cmethoxybenzaldehyde. Catalytic hydrogenation completed a synthesis of munduseran 108; unfortunately, no method of oxidizing this compound to the rotenoid
Derric acid 105
Me0\ MeO/
OH
0 C H : O
108
Munduserone 109
munduserone 109 has yet been found. Of interest is the specificity in the direction of enolization and in the point of condensation. Conjugation with the benzene ring rather than the cyclic oxygen atom seems to control these reactions. Here again DCCD has provided most striking improvements. In the form of its pytrolidine enamine, dimethoxychroman-3-one 106 reacts with tubaic
506
The Synthesis of Oxygen Ring Compounds
acid and DCCD giving dehydrorotenone directly;e6and if the acid chloride 110 of tubaic acid is used with the enamine 111, the yield of dehydrorotenone 112 runs as high as 15 % instead of the disheartening 0.0023 % for equivalent stages in the original method utilizing Hoesch condensations. Again an important feature is the avoidance of acids that would disrupt the propenyl grouping of rotenone.
Tubaic acid Chloride
111
110
OMe
Dehydrorotenone 112
a.
The thermal condensation techniques explored by Mentzer and his colleaguesg7provide another variation of this route that has not yet been evaluated fully. The essential points can be summarized schematically:
-OF1
m E , t +
\ 113
0
\
'14
C0,Et
4.
The Rotenoids
507
Phloroglucinol can replace resorcinol but, as always in this technique, it is best to etherify all hydroxy groups not actually needed for the reaction. A multiplicity of free hydroxy groups allows side reactions and tends to favor condensation modes leading to 2-pyrones (coumarins) at the expense of the required 4-pyrones. Another trouble is that the isomeric 3-0x0-esters 113 and 114 cannot be separated easily so a mixture is used and each component gives its own series of condensation products. Originally suggested by taxonomic considerations and later established beyond doubt by experiments with tracers, the idea that the rotenoids are, biosynthetically, modified and elaborated isoflavones has stimulated a number of efforts to achieve a laboratory synthesis along similar lines. The chemistry of elliptone offers several examples. The first approach was to obtain a 2bromornethylisoflavone containing a free hydroxy group at the 2’-position suitable for an internal etherification. Radical-catalyzed bromination of 2methylisoflavones does yield the desired 2-bromomethyl compoundses but because in most cases the protection of sensitive substituents poses formidable problems we here illustrate a more general method reported by Mehta and Seshadri :0°
The product has been further elaborated by Kawase and Numata,lo0 who used the Duff condensation to introduce a formyl substituent (the question of orientation is considered shortly) and the bromomalonate method to complete a furan ring as shown overleaf.
508
The Synthesis of Oxygen Ring Compounds
H
O
Eto2c& C0,Et
W BrHC(C0 El),
A \
\
1
-OH; heat
1
CO,H
4 2 ? g E
% 0 \
This furan is a close relative of the rotenoid elliptone 115a.The next synthesis from isoflavone precursors illustrates a superior method of providing for fused furan rings.101 The Claisen migration in 116 is specific in direction and gives 117 because the transition state (or intermediate) represented by 118a is less cross-conjugated and therefore energetically more favorable than the alternative 118b needed for migration in the opposite direction. The same influences control similar migrations (and electrophilic
O W \OMc OMC
Elliptone 115a
5) %e
OMc OMe
om 4. The Rotenoids
0
11811
118b
509
0
substitutions) whether the second ring is 4-pyrone (as here), 2-pyrone, furan, or benzene, or even not a ring but certain substituents, and consequently it is usually difficult to construct with elegance compounds having annelation of the kind seen in isoelliptone 115b. One solution is to work with nonaromatic systems and aromatize them at a late stage. Subsequent steps in elliptone synthesis are shown overleaf. An interesting feature is the selective demethylation of the 2’-niethoxy group by aluminum chloride in refluxing methyl cyanide.Io2 Other solvents are not satisfactory. Presumably the aluminum chloride forms a pyrylium salt with the pyrone grouping and the aluminum atom is then near enough to the 2‘-methoxy group to coordinate with it and promote its demethylation, as indicated in diagram 119. Recently, boron tribromide has been used for the same The final steps (not shown) are the acetic anhydride-sodium acetate cyclization of elliptic acid 120 to the chromenochromone, dehydroelliptone, and the reduction of this to (&)-elliptone by the borohydride method. No resolution has been reported. Racemic forms of isoelliptone 115b and of munduserone 92 have been obtained similarly and no doubt the method is of considerable applicability. Notwithstanding that, the method does not solve the problem convincingly since the isoflavone nucleus is used merely as a protective device for one hydroxyl group and is destroyed and has to be reconstituted. (In contrast, similar sequences form an excellent ingress to pterocarpans.) This deficiency has recently been made good by application of sulfur ylide chemistry.
510
The Synthesis of Oxygen Ring Compounds
I
-*o&l
CH:O
2) IlOSO, Na10,
. . - o a\ ]
117
AICI,
l
Elliptic acid
OMe
120
When dimethylsulfoxonium methylide adds to the 2,3-double bond of isoflavones it initiates a variety of reactions.lo4*1O6If an excess of reagent is avoided the result is simple, a cyclopropane derivative being formed : OMe
Me
I
4. The Rotenoids
511
But if there is a free 2'-hydroxy group, a series of ring openings and closures leads to a vinylcoumaranone system, perhaps as follows, although this scheme allocates no particular role to the important 2'-hydroxy group itself:
I
At first sight such vinylcoumaranones seem to be a long way from the desired goal, yet they are actually isomeric with the corresponding chromanochromanones and when heated with pyridine at 100" for 48 hours can be converted into them in high yields (-80 %) as illustrated for isorotenoneio6 121:
Isorotenone 121
OMe
512
The Synthesis of Oxygen Ring Compounds
There is still one problem to be solved. In the foregoing elegant method the sulfur ylide provides what eventually becomes the cyclic methylene group. In nature, however, the cyclic methylene group is perhaps provided by attaching the methyl group of the 2'-methoxy group to position 2 in the isoflavone nucleus. No comparable cyclization has been achieved in the rotenoid field, though something like it is known elsewhere. Isoflavones are prerequisite for much of the work among rotenoids. They are generally obtained from phenol and phenylacetic acid derivatives by connecting these together in Friedel-Crafts style and effecting cyclization by Claisen condensations with ethyl oxalate or with ethyl orthoformate and piperidinelo7 (the most popular method nowadays), though others are available. The thallium 111 oxidation of chalconeslOBis a different approach which was of additional interest when it was thought to simulate a biqsynthetical route to isoflavones:
fPh
An amusing isoflavone synthesis is provided by the following condensation of a chromanone with a quinone,'OS with its repeated prototropic shifts. The product is related to neotenone and the yam bean rotenoids.
Me0
4.
The Rotenoids
513
Another way of reaching interesting isoflavones is an unusual variation of the Claisen ally1 rearrangement applied in this example to an ether made from umtatin, a natural 2-hydroxychromone with the furanoid system found in rotenonello :
5. COUMESTANS AND PTEROCARPANS
Coumestan is the name of the fundamental ring system 122a characteristic of a small, well-defined group of natural products related to the isoflavones. Pterocarpan describes the system 122b, a reduced form of 122a. The numbering of pterocarpan has been generally agreed for some time and is adhered to, but several systems have been used for coumestan and the one shown is
514
The Synthesis of Oxygen Ring Compounds
perhaps the most common. Even the name “coumestan” itself is unsatisfactory, and it is surprising that so few authors prefer a more systematic name,* for example, benzofurano(3‘,2‘: 3,4)coumarin.
8
2 \
11
%3
\10
@
Pterocarpan 122b
Couniestan 122a
It is convenient to begin with syntheses of coumestan derivatives, that of wedelolactone 123 illustrating the first success.111J12As with so many syntheses in the isoflavone field, this commences with a deoxybenzoin 124, sodium powder and ethyl carbonate effecting a Claisen ester condensation and cyclization occurring spontaneously to give, after enolization, a 4hydroxycournarin 125 according to the technique introduced by RobertThe acidity of such coumarins is approximately that of acetic acid and ensures the production of the salt 126, which is readily isolated as such. Various workers113”have described minor modifications of this method, while Deschamps-Vallet and Mentzerllsb have shown how basic media can be dispensed with altogether if thermal condensation techniques are employed. The arrows show that the salt, notwithstanding its lactonic nature, is stable
Wedelolactone 123
126
124
/ El .I CO,:
125
* Chernicul Abstracts uses 6H-benzofuro(3,2-c][I I-benzopyran-6-one. with numbering as for pterocarpan above.
5. Coumestans and Pterocarpans
515
to nucleophilic attack and therefore to the large excess of base used for the reaction. Heated with hydrogen bromide, the 4-hydroxycoumarin 125 cyclizes to the tetramethoxycoumestan 127 directly, so a selective demethylation of one methoxyl group must occur. If a guess may be hazarded, the known basicity of 4-hydroxycoumarins would suggest formation of a pyrylium salt in which the arrangements for demethylation indicated in 128 seem particularly suitable. Careful treatment with hydrogen iodide composes the last step112 in which 127 yields the monomethyl ether 123. The interaction shown in 127 tends to protect the one methoxy group from protonation and hence from demethylation by iodide ion; other effects that probably promote demethylation at other positions are discussed on pp. 538, 540.
Another point worth attention is that the wedelolactone kernel is indifferent to mineral acid despite the presence of a benzofuran nucleus which is usually easily protonated and may then undergo various reactions such as ringopening with water giving a phenylacetaldehyde derivative rapidly resinified by acids. In wedelolactone, defense against such attacks is provided by the pyrone ring, which, as noted previously, is relatively basic. It can be assumed that the first proton is neutralized as the salt 129; at this point the benzofuran oxygen atom is unable to participate in attaching any further proton except in extreme conditions. l~~~ Coumestrol 130 has been synthesized similarly, as have l u ~ e r n o 131 and derivatives of psoralidinells 132, sativo1114133, and others,”6 with an improvement in the use of the more reactive ethoxycarbonyl chloride instead of ethyl carbonate for preparing the 4-hydroxycoumarin derivatives, and of anilinium chloride for demethylation. Coumestrol is of interest because it has estrogenic activity (it contains a 4,4’-dihydroxystilbene nucleus). The more highly substituted coumestans appear to have little or none, but several
516
The Synthesis of Oxygen Ring Compounds
have activity against plant pathogens and may be produced in response t o fungal attack, and so on, and therefore can be classed as phytoalexins.
Lucernol 131
Coumestrol 130
,AS \
Psoralidine 132
kc'o-
OMe
\
OH
Sativol 133
These physiological activities have intensified interest in the coumestans and in their synthesis. Two elegant and powerful methods, one devised by Wanslick and one by Jurd, may be regarded as having replaced the method outlined. Wanslick, Gritsky, and Heidepriem1I7 demonstrated that oxidation of a mixture of 4,5-dihydroxy-7-methoxycoumarinand catechol with potassium iodate in acetate buffer led to wedelolactone in 85 % yield:
Wedelolactone
Enols other than 4-hydroxycoumarins can be used,l17 and examples have been described involving groups as sensitive as the amino group.118 As to mechanism, one might exclude radical coupling on the grounds that high yields of unsymmetrical products could never be obtained thus. Since for one cornponent a I ,Zdihydroxybenzene group is essential, 'nascent' I ,2-quinones may be intermediates116a, the anion of the hydroxycoumarin adding to them; and with a simple variation on the same theme there is no difficulty in accounting for the remarkably ready cyclization that completes the sequence. The
5. Coumestans and Pterocarpans
517
method is necessarily restricted to derivatives of 1 1,I2-dihydroxycoumestan but has been used very successfully for numbers of often with ferricyanide or hypoiodite as an alternative oxidant.
Originally studied in the hope of elucidating the structure of anthocyanins, the hydrogen peroxide oxidation of flavylium salts is a complex matter. The result for acetic acid solutions is merely a ring fission:1a3
R = Ph, OMe, or OC,H,,O, In aqueous methanol pH 5-7, however, the oxidation readily affords compounds derived from 2-phenylben~ofuran,~** other studies conducted by Jurd12, suggesting the following course of events:
-3
-3
qc’:i 0
R
0
t
518
The Synthesis of Oxygen Ring Compounds
Applied to 3-methoxyflavylium salts containing a 2'-hydroxy group, the oxidation gives first an ester, but this usually rapidly cyclizes to a coumestan derivative :124*128
Coumestrol
Since flavylium salts are readily available in considerable variety, the method is a powerful and flexible one. The preceding synthesis supplies coumestrol in 50% yield.124 Syntheses of medicago112Tare useful for comparing this technique with the Wanslick coupling reaction for obtaining certain substitution patterns. The coupling readily gives the trihydroxycoumestan 134, but attempts at the selective methylenation (by CH,I,) required to give medicagol 135 afford, at best, only 5 % of this compound.'22 For the flavylium salt route, the components 136 and 137 are needed; these condense very readily to form the salt 138, which is debenzylated by hot aqueous acid and oxidized to form the coumestan. The oxidation affords a yield of about 40%, but of course the '0% \
' / O H O
\
OH
134
Medicagol 135
I'll-()
OMc
137
136
138
5.
Coumestans and Pterocarpans
519
somewhat elaborate nature of the starting materials may be thought to detract from this advantage. In general, Wanslick coupling offers the better route to derivatives of 1 I , 12dihydroxycoumestan (which are numerous) but is not adaptable to other hydroxylation patterns, an area where Jurd’s flavylium salt oxidation excels.1aB*120 Of course the scope of both methods is greatly increased by the proper use of protective devices as will be seen in the following examples, the first of which is a very much improved synthesis of medicagol 135 utilizing Wanslick oxidation
Trifoliol 139 presents a difficult problem elegantly solved by flavylium salt oxidation. The solution depends upon the ready availability of the benzoate 140, which permits preparation of the flavylium salt 141 and thence the benzofuran 142, which is methylated before the protective groups are removed.131 7-Hydroxy-l l ,1Zdimethoxycoumestan 143, which occurs in alfalfa, can be made132from the easily accessible 7,11,12-trihydroxycoumestanby means of a characteristic use of the relatively uncommon protective reagent, dichlorodiphenylmethane :193
I ) PhCOCllpy
2) HCllAcOH
143
\ r-
a
J
520
5.
Coumestans and Pterocarpans
521
Wanslick coupling is usually the method of choice if strongly acidic media have to be avoided, as in synthesizing erosnin120144 with its benzofuran moiety :
( o m / -----, o \
OH
Erosnin 144
With the knowledge that coumestans are biogenetically isoflavanoids and not true coumarins, Grisebach and his colleagues134have investigated the formation of coumestrol 100 from 2',4',7-trimethoxyisoflavanone 145. This material is heated at about 200" with pyridine hydrochloride in air, and a tiny amount (-0.2 %) of coumestrol results. Almost certainly demethylation allows ring closure to a dehydropterocarpan derivative 146, most of which would be destroyed because of uncontrollable oxidation of the phenolic rings, but some of which is oxidized at the methylene group to the coumarin. This is less sensitive to oxidation because of the electron-withdrawing effect of the carbonyl group, and a little survives. Radical oxidation (by air) of ring methylene groups in dibenzopyrans and related systems is known, having been investigated by Robertson and his c o l l e a g ~ e s , and ' ~ ~ it may be significant that 146 has in fact been isolated during on the heartwood constituents of Swartzia nradugascariensis.
With these considerations we find ourselves discussing pterocarpan chemistry. There are but two independent routes to pterocarpans, and one of these leads through coumestan country. A case in point is the synthesis of maackiain, which begins with the benzyloxydihydroxycoumestan 147
522
The Synthesis of Oxygen Ring Compounds
intermediate in one synthesis of medicagol. From here, Fukui and his coworkers130 effected methylenation and proceeded by lithium aluminum hydride reduction to the alcohol 148, which they cyclized by dissolving it in boiling diethylene glycol. This is the standard method and gives very satisfactory yields of what are known as dehydropterocarpans. Usually, of course, such an etherification is effected using acid catalysts capable of producing carbonium ions from the allylic alcohol segment; here, however, acids have to be avoided on account of the sensitive benzofuran nucleus. Saturation of the olefinic bond in the dehydroptercarpan 149 is not easy but can be accomplished catalytically with palladium-on-charcoal at 70 atm and 70". Rhodium catalysts are sometimes preferred. In the present example the benzyl group is lost simultaneously and a racemate corresponding to the natural product maackiain 150 is produced. It has not been resolved (none of the synthetic pterocarpans have been). The catalytic method ensures that the ring junction is cis as in all members of this series. Other ptercarpans obtained similarly
147
Maackiain 150
149
include 3-hydro~y-8,9-dimethoxypterocarpan,~~~ 3,8,9-trimethoxypterocarpan,138 ( f)-pterocarpinlZ0 151, and (f)-4-rnethoxyptero~arpin.~~* In a slightly modified version of this synthesis of 4-methoxypterocarpin 152, diborane in tetrahydrofuran was used to reduce the coumarin carbonyl group to methylene without opening the lactone ring.119 The yield is -50 %
5.
Coumestans and Pterocarpans
523
Pterocarpin 151
152
and theidea worthimitation. (Anotherexample occurs in Section 6.) Curiously, simpler coumarins do not behave in such a straightforward fashion. Coumarin itself gives 2-allylphen0l.~~~ The other route to pterocarpan derivatives, which requires isoflavones as starting materials, was explored first by Suginome and Iwadare.lao Another synthesis of maackiainlq1 153 is offered in illustration of the basis of the method :
524
The Synthesis of Oxygen Ring Compounds
The reduction of isoflavones by borohydride has been discussed previously. The subsequent cyclization is conducted in hot 50% acetic acid so probably involves a benzylic carbonium ion as shown. That the requisite cis ring fusion results is merely a consequence of ring strain, which is somewhat greater in inermin, ~ the trans fused arrangement. Racemic forms of p t e r o ~ a r p i n , ' ~ 8-methoxyhomopterocarpin,143and other pterocarpans including analogs of n e ~ d u l i n have ' ~ ~ been made by this method without significant variations. In all these examples a 2'-hydroxy group must be provided for in the isoflavone precursor, often by selective demethylation just as in some routes to the rotenoids as described earlier. Protective techniques are also commonly used, and combined with a new method of obtaining isoflavones from enamine condensations they form a very flexible ingress to pterocarpan chemistry. An application to (&)pterocarpin outlines the method
3) AcOH
Pterocarpin
No attempt seems to have been made to synthesize trifolirhizin, the 8glucoside of 3-hydroxy-8,9-methylenedioxypterocarpan (inermin), although this would be of special interest in that it would probably also constitute a method of resolution in this series. The last compound discussed here is pisatin 154, curious in its hydroxy substituent, unique in the series, and in having rotation of the opposite sign to pterocarpin, from which it is formally derived. Pisatin is a phytoalexin of
6. Amphipyrones (Pyronoquinones)
525
the common garden pea, whereas pterocarpin comes from a tropical heartwood-the plants concerned, however, both come from the same family Leguminosae, and even the same subfamily, the Papillionaceae. (&)-Pisatin has been synthesized, since it can be obtained from pterocarpin. The method requires a strictly controlled acid hydrolysis of the benzylic ether linkage in pterocarpin, ION hydrochloric acid in ethanol being the catalyst. About 40% of the flavene 155 is obtainable, and with osmium tetroxide affords the black osmate 156, which is decomposed by sodium carbonate in mannitol. The cyclic body is formed at once, notwithstanding the absence of the usual acid catalyst, and it is thought that a nucleophilic displacement occurs as the osmate collapses, as indicated.’4EDisplacements at the benzylic position are usually very easily effected.
155
Pterocarpin
Pisatin
154
BJ 156
6. AMPHIPYRONES (PYRONOQUINONES)
The number of fungal metabolites containing or derived from nucleus 157 is steadily increasing.14’ This nucleus is often called the “pyronoquinone” nucleus, but the name is unfortunate in that it suggests a fusion of quinone and pyrone rings instead of the elision characteristic of 157. The name “amphipyrone” is offered here as a substitute that indicates the correct number of oxygen atoms and their location at opposite sides of the ring system. At a time when none of these compounds had been allocated a * structure the name “azaphilone” was coined for them because of their most striking reaction in which ammonia rapidly replaces an oxygen atom by the imino group. One of the first members to be extensively investigated was
526
The Synthesis of Oxygen Ring Compounds
~ c l e r o t i o r i n(158), ~ ~ ~ a yellow pigment of Penicillium sclerotiorum van Beyma and other fungi, which with ammonia yields the red sclerotioramine 159.
157
Sclerotiorin 158
Sclerotiorarnine 159
Compounds with nucleus 157 are sensitive in other ways also, especially to nucleophiles, and their syntheses are correspondingly difficult. Only one reasonably simple synthesis of the system has been achieved, and although it has not yet been applied in the case of a natural product it is discussed here because of its intrinsic interest and especially because it does in fact constitute the extension of a simple pyrone. Yamamura, Kato, and Hiratalas heated 2,6-dimethyl-4-pyrone (160) with cyanoacetic acid in a large excess of acetic anhydride and obtained the extended pyrone 161. There is limited precedent for such a condensation at the carbonyl group of a p y r ~ n e , though * ~ ~ these groups are usually virtually inert because of interactions indicated in 160. The conditions suggest that (a) cyanacetic acid condenses with itself under the influence of acetic anhydride giving the dicyanoketone 162, then (b) the enol of the dicyanoketone attacks position 4 in the pyrone whenever this is activated by protonation as in diagram 163. Several other, rather similar schemes can be envisaged. However the reaction occurs, it gives the relatively good yield of 30%, and conversion into the diamide 164 is not difficult to accomplish with sulfuric acid. But what makes the method attractive is that polyphosphoric acid converts the extended pyrone 161 directly into the cyclized amphipyrone 165 in quantitative yield. Direct substitution into a pyrone ring is very rare, especially in acidic media where most pyrones would be expected to exist in a protonated form as pyrylium salts that are no more susceptible to electrophilic attack than are pyridinium salts in a like situation. Since the immediate substitution-cyclizationproduct may well have structure 166, it is reasonable
6. Amphlpyrones (Pyronoquinones)
527
to account for the special behavior in terms of the large increase in aromaticity, the nucleus of ion 166 being isoelectronic with naphthalene. This aromaticity is lost if a proton is removed and the amino group then becomes an enamine function, acid hydrolysis of which yields the enol 165. With ammonia, 165 undergoes the characteristic replacement giving the isoquinoline derivative 167.
161
164
H ,NCO
CN
HoI 0
165
CONH,
'.$" H
H2N&Y ' / C'ONH,
I
CONH, 0
OH 166
LONH,
0 167
To date, however, all syntheses of the natural products have been planned along lines based on phenol chemistry and elaborated by Whalley and his colleagues who, taking up an earlier idea,14Bbegan'51 with a ranging shot fired at the relatively stable compound 168, tetrahydrosclerotioramine, instead of the much more sensitive parent compound, sclerotiorin 158. The phenolic ketone 169 was prepared by allowing the cadmium derivative of (+)-3,5(S)-dimethylheptyl bromide to act on 3,5-dimethoxy-ll-methylphenylacetyl chloride and then demethylating the product. Usually the Gattermann aldehyde synthesis with hydrogen cyanide and hydrogen chloride converts resorcinols into aldimines, which are then hydrolyzed, but in this case the imine 170 inevitably cyclizes to the isoquinoline 171. Chlorination with sulfuryl chloride in acetic acid gives 172, and the last step is
528
The Synthesis of Oxygen Ring Compounds
acetoxylation with lead(1V) acetate, a reaction extensively explored by von W e s ~ e l e y . 'Of ~ ~course, this reagent is far from specific either positionally or stcreochemically, and 168 is produced along with other compounds including the epimeric acetate, from which it seems not to have been separated. Nevertheless, the result establishes the met hod.
oh
Ac 0
H
HO
H
0 168
169
171 R = H 172 R = CI
170
A different method of producing an amphipyrone appears in a synthesis, by Galbraith and Whalley,lS3 of (f)-ascochitine 173, a metabolite of Ascocliyra yisi Lib. and A. fabae Speg. and one of the simplest of the natural amphipyrones. Methylation of 3,5-dimethoxybenzyl cyanide with sodium in liquid ammonia followed by iodomethane gave the starting material 174, which was converted by standard methods into the phenolic ketone 175.
Ascochitine 173
bl C
O
I74
HO 175 R = H 176 R = CO,H
The carboxyl group was introduced between the hydroxy substituents by treatment at 180" with potassium hydrogen carbonate in glycerol, this
6.
Amphipyrones (Pyronoquinones)
529
orientation being usual where there is a kubstituent bulky enough to shield positions 4 and 6. Heated with ethyl orthoformate and an acid catalyst (hydrogen chloride and toluene-p-sulfonic acid are both useful) the acid 176 gave ascochitine 173 in one step, though of course several stages must be involved-perhaps cyclization and other steps occur as indicated in diagrams 177 and 178.
t 1 1
177
178
For other syntheses a return was made to the acetoxylation technique. Beginning again with phenolic ketone 169, reaction with ethyl orthoformate affords the aldehyde 179, and sulfuryl chloride (catalyzed by benzoyl peroxide) then gives the halogen derivative 180. Here enolization, hemiacetal formation, and dehydration should give the amphipyrone 181, but this seemed far too sensitive to be manipulated directly so it was used in the protonated form 182 obtained merely by treating the aldehyde 180with hydrogen chloride in ether. As before, lead(1V) acetate supplied a mixture from which it was possible to isolate a mixture of tetrahydrosclerotiorin 183 and its epimer.
179 R = H 180 R = CI
181
0 182
Tetrahydrosclerotiorin 183
530
The Synthesis of Oxygen Ring Compounds
These have not yet been separated, but the properties of the mixture leave no doubt as to the identity.ls4 A mixture of sclerotiorin epimers has been obtained similarly.1ssStandard methods suffice to convert acid 184 through the acid chloride and diazoketone into the bromomethyl ketone 185, which is then debenzylated using boron tribromide and acetylated. The product 186 is transformed into a phosphorane under very mild conditions of minimal basicity by treatment with triphenylphosphine and methyloxirane at 25" for 16 hours. After removal of2-bromopropan- 1-01 the mixture is allowed to react with (f)-2,4-dimethylhexrrans-2-enal at 130" under nitrogen, and finally the acetate groups are removed by alkaline hydrolysis under nitrogen. This sequence would be expected to yield the phenolic ketone 187, but the compound obtained is said to be a hydrate, which, rather surprisingly, is formulated as 187a.
I> ''-0Q??m2
0-Ph 184
187
H
R
O
m Br
OR 185 186
R = CH,Ph R = AC
187a
The next steps are the now familiar ones where ethyl orthoformate introduces an aldehyde function and sulfuryl chloride a chlorine atom, the product being 188a; a new aspect is the use of phosphorus pentoxide to induce formation of the amphiquinone 189 as a racemate corresponding to (+)aposclerotiorin, a reduction product of the natural pigment. Natural (+)- aposclerotiorin formsa useful relay at this point,sinceevenunder nucleophilic attack as mild as that of acetate ion its heterocyclic ring opens to form the ketonic aldehyde 188a, thus demonstrating the lability of the amphipyrone nucleus. Schematically, the fission takes the path shown in 188b and 188c. Acetoxylation of the natural compound 189 generates a mixture, and again a fraction can be isolated that contains both (+)-sclerotiorin 158 and an epimer too similar to be readily separated. The difficulty is always the same: there are two asymmetric centers only, but they are too far apart to have any
6. Amphipyrones (Pyronoquinones)
531
influence on each other and too far for much hope of achieving asymmetric synthesis by induction.
tl0
HO 188a
Aposclerotiorin 189
188b
l88c
A further interesting elaboration is possible. Ethanolysis of (+)-sclerotiorin with sodium ethoxide at 0" gives the alcohol 190, seemingly without collapse of the heterocyclic ring. It will be appreciated that now nucleophilic attack no longer leads to direct aromatisation as in 188b. Thealcohol reacts with diketene in hot pyridine-benzene in the presence of hydrogen chloride giving (presumably) first the acetoacetic ester 191 and then the cyclization product 192. The yield is good, and the methodlS5should offer a promising doorway to syntheses of the natural analogs*5Esuch as rotiorin, rubropunctatinl57 and monascorubrin15*193. O
HO
Y\\\
G
\o
0
0
190
0
191
O=CC5HI
I
0 Monascorubrin 193
532
The Synthesis of Oxygen Ring Compounds
Finally, a synthesis'59 of (f)-mitorubrin 194 illustrates some minor variations. The appropriate phenolic ketone with ethyl orthoformate and hydrogen chloride rapidly gives the pyrylium salt 195, which is precipitated at once with ether. With ethanolic potassium acetate it immediately yields the ketonic aldehyde 196. Since the amphipyrone is too unstable to withstand isolation, a solution of the aldehyde 196 in ethanol containing phosphorus pentoxide is used to obtain it in solution and it is then treated in situ with lead(1V) acetate to produce the more stable compound, 197. The acetyl group is removed by sodium ethoxide and the alcohol esterified by 2,4dibenzyloxy-6-methylbenzoicacid in the presence of trifluoroacetic anhydride. Then debenzylation with boron chloride at - 70" gives (f)-mitorubrin 194. The natural compound is the (-)-isomer and can be extracted from the phy t otoxic fungus ,Penicillium rubrunt.IB0 0
'vow H
r
ti0
194
H
195
Y \
OH
CHO 196
10,
AcO
0
197
I n one sense, amphipyrones are derived from isochromene and are consequently related to numerous compounds of the isocoumarin type. Most of these compounds contain true benzene rings and are therefore not of immediate interest here; but two others, citrinin and fuscin, both fungal metabolites. are lionaromatic and have been synthesized by methods indicative of the extent to which the quinonoid part of the system can be created by the oxidative techniques ignored so far. The synthesis of citrinin 198 begins with the resorcinol derivative 199, which, in KolbC conditions, suffers carboxylation to the acid 200, the formation of the isomer being prevented by the bulkily branched side chain. The Gattermann reaction161is often less sensitive than carboxylation to steric effects, so there is no difficulty in producing the aldehyde 201. Finally, dehydration by sulfuric acid yields citrinin 198 as a racemate that can be resolved as the brucine salt.162
6. Amphipyrones (Pyronoquinones)
Citrinin 198
201
533
199 R = H 200 R = CO,H
202
Dihydrocitrinone 203
The synthesis of dihydrocitrinin 202 is even easier, since formaldehyde condenses very readily with the methyl ester of acid 200 and the product cyclizes at once.la2The gentlest of dehydrogenations (bromine or mercuric oxide) now gives the quinonoid citrinin.ls3 The idea appears in various forms in isochromene chemistry. For example, Curtiss, Hassall, and N a z a F discovered that a fungal mutant no longer able to synthesize citrinin generated dihydrocitrinone 203 instead, and they used diborane to reduce this to dihydrocitrinin. In this work oxidation to citrinin was accomplished with manganese dioxide or simply with air. In strong contrast, the removal of further hydrogen to give a true amphipyrone seems very difficult to achieve without causing a deeper disruption-at all events, no such dehydrogenation has yet been described. Fuscin 204 is another fungal metabolite. The description of amphipyrone can hardly be applied to this compound; however, in addition to its extended quinonoid system it does possess two modified pyran rings. The synthesis'e5starts with a standard chroman synthesis, methyl 3,4,5-trimethoxyphenylacetate and 3-methylbut-2-enoyl chloride giving the chromanone 205 and Clemmensen reduction the chroman 206. Another acylation introduces an acetyl group, and borohydride reduction produces the corresponding alcohol which at once lactonizes, giving 207. When demethylated, the product was simply left in air to suffer oxidation to the quinonoid fuscin 204. In theory, the molecule could exist in any one of several tautomeric forms. That only 204 is observed is a consequence of the para quinonoid nature of this arrangement (the other possibilities are ortho quinonoid) and of the interactions indicated between certain parts of the molecule (such interactions arc reduced or lost in tautomers).
534
The Synthesis of Oxygen Ring Compounds
Fuscin 204
206
205
207
7. XANTHONES *
Xanthone occurs naturally as derivatives in a few families of higher plants, notably the Guttiferae and the Gentianaceae. Such compounds contain from one to five hydroxy and/or methoxy Sometimes C, isoprenoid side chains make an appearance. A few fungal xanthones are also known, and among these C-methyl groups and chlorine substituents are found.l66 The numbering of xanthone is now always as shown; letters are used in biosynthetical A referring to the acetate and B to the shikimatederived ring.
Although xanthone contains only four different locations for substituents, these being duplicated symmetrically, even with just five hydroxy and/or methoxy groups to dispose around the periphery one is faced with some hundreds of possibilities. About 70 of these are currently known as occurring
* The author is grateful to Dr. H.
with the preparation of this section.
D. Locksley and Dr. F. Scheinmann for their help
7.
Xanthones
535
in plants, the great majority having been discovered since the last extensive review^^^**^^^ were published around 1960. Clearly, selectivity and orientation control are the synthetical problems. It simplifies matters somewhat that while each benzene ring in xanthone interacts strongly with the central pyrone nucleus, there seems little transmission of effects across this nucleus from one benzene ring to the other. And it is a decided advantage that the xanthone nucleus is normally stable both to bases and to acids. The stronger acids merely protonate the carbonyl oxygen atom (reversibly) giving hydroxyxanthylium salts,, and ordinarily there is no danger of Wessely-Moser
rearrangement^.'^^
At present the xanthone synthesis introduced by Grover, Shah, and Shah170enjoys by far the greatest p0pularity.1~1-~~0 It requires simple, usually accessible, materials: a salicylic acid derivative and a suitable phenol. It utilizes the simplest of techniques: the components are merely heated together with zinc chloride in phosphoryl chloride as solvent. In the following example the xanthone 208 formed by this method needs only selective methylation by diazomethane to yield lichexanthone (209), a metabolite of the lichen Parnielia formosana:
208 R = H; Norlichexanthone 209 R = Me; Lichexanthone
?
CHINl
Reaction times vary from 0.5 to 24 hours. Sometimes the solvent is omitted and the temperature raised to 180" as in the Nencki reaction,lTBand sometimes polyphosphoric acid is preferred to the zinc chloride mix.174.176 Some acids (phloroglucinol carboxylic acid, resorcinol-2-carboxylic acid) are easily decarboxylated and require the mildest conditions for success, but gentisic acid (2,5-dihydroxybenzoic acid) provides great difficulty. It fails to react, which is a nuisance because quinol nuclei are common amongxanthones and quinol itself is too resistant to electrophilic substitution to be used. Curiously, 2-hydroxy-5-methoxybenzoicacid is perfectly satisfactory in Grover-Shah-Shah (GSS) condition^,^^^.^^^*^^^ and the next example is of its (210): use in a preparation of I ,3-dihydroxy-7-metho~yxanthone~~~
210
536
The Synthesis of Oxygen Ring Compounds
This striking effect of methylation has not been explaincd. It may be connected with the suppression of ionization in the acylium ion 211, which is presumably the intermediate in what amount to Friedel-Crafts conditions. The first ionization would give a ketene 212 (there is analogy181), and a second would now greatly reduce the electrophilic character of the carbonyl group, as indicated:
21 1
212
It follows, therefore, that free hydroxy groups at positions 4 and 6 in the salicylic acid would have no comparable effect and that a 3-hydroxy group would interfere but to a smaller extent; in all cases methyl ethers would give better results than the phenols do. No proper study of this kind has been undertaken, but qualitative comparisons appear to bear out these conclusions. Strictly speaking, it is not xanthones that are usually formed in the general GSS reaction, but 2,2'-dihydroxybenzophenonesthat have to be cyclized in a separate step. Heating to about 200" in water in a sealed tube or an autoclave has been recommended, although lower temperatures suffice with acid catalysts:
Xanthones are produced directly only if the benzophenone intcrmcdiate carries another hydroxy group at the 6 or the 6'-position, that is, if an alternative site for cyclization is available.170The following diagram 213 suggests that coordination with zinc ion might usually impede cyclization by forcing the molecule into an unfavorable conformation with the two hydroxy groups far apart; when the third hydroxy group is present cyclization is still possible-indeed, promoted, since the conformational restriction now insists that two hydroxy groups be mutually accessible as in diagram 214.
7. Xanthones
213
537
214
An important deviation from the general orientation rules is found in condensations with orcinol, the incoming carbonyl group becoming attached to the point between the hydroxy groups of this particular resorcinol derivative. Such behavior is common in the orcinol series and is to be attributed to the difficulty of pushing a bulky group (carbonyl coordinated to zinc ion) between one hydroxy group and the relatively large methyl group. Thus 8resorcylic acid and orcinol yield xanthone 215 directly, whereas orsellinic acid 216 and resorcinol yield a benzophenone and then the xanthonelTO217.
These idiosyncracies aside, the GSS reaction provides a wide variety of the simpler polyhydroxyxanthones and their ethers where these are not unduly sensitive to acids. The methoxy group (and methylenedioxy groupl7@)is usually stable unless it arrives at position I (or 8) in the final xanthone, in which case it tends to be lost as it would be from any other o-alkoxycarbonyl compound. Mixtures may result, and demethylation can be completed, if ~ ~ - ~route177 ~ ~ to desired, by aluminum chloride or boron ~ h l o r i d e .A~ good xanthone 218 uses aluminum chloride in the GSS melt itself to open the way to
The Synthesis of Oxygen Ring Compounds
538
xanthone formation and then to secure free hydroxy groups at position 1 :
0
L
0
218
That at higher temperatures other methoxy groups may be affected is one of the defects of the older technique (Nencki reaction) in which the components were simply strongly heated with zinc chloride alone. With resorcinol derivatives differences in orientation result, ~ o o presumably , ~ ~because ~ ~ at higher temperatures the products are determined thermodynamically rather than kinetically. Thus resorcinol and 2-hydroxy-5-methoxybenzoic acid under GSS conditions yield xanthone 219, whereas under Nencki conditions the resorcinol is acylated at the 2-position so that, with concomitant demethylation, the product is the well known pigment euxanthone 220.
HO \
0 219
\
0
OtI
Euxanlhone 220
An important demethylation occurs where pyrogallol nuclei are con~ e r n e d . ' ~ Zinc ~ . ' ~chloride ~ under GSS conditions converts pyrogallol trimethyl ether into 2,6-dimethoxyphenol in high yield. Sulfuric acid has the same effect, and the reaction often occurs in preference to the better known demethylation at position 1. As illustrations syntheses of xanthones 221 and 222 should suffice (as shown on top of facing page). This selectivity is often a valuable feature for which buttress effects1B4-186 have been held responsible,180since a group flanked on both sides by other (large) groups may be pushed out of the plane of the benzene ring and its behavior modified accordingly. Further examples will appear shortly.
~
~
7. Xanlhones
539
Me0 221
OH M Me e 0O
m
O
M
e
:;2s.
~
\
\
"
.
O
q
p
M \
\
e 222
OMe OMe 0 0 However, this explanation may be accepted only cautiously at the moment, for there is a dearth of examples in which flanking groups other than methoxy groups have been studied. A possible alternative view assumes that coordination of metal ions or hydrogen bonding phenomena are responsible, the central methoxy group being always subject to a twofold attack, the outer ones to not more than a single attack. The idea is sketched in 223, while a pleasing double application of the reaction has been described by Gottlieb and his colleagues,18swho converted the ether 224 into the phenol 225, which occurs in some species of Kiefmeyera. Me 0....Z"++
h& 1
&em ,:Zn++
223
@:Le \
0
224 R = M e ]H&Q 225 R = H
A different approach to xanthone synthesis is based on the cyclization of an o-phenoxybenzoic acid. The diphenyl ether link is forged by the Ullmann reaction as in a synthesisls6 of 2-methoxyxanthone 226, used for identification of 2-hydroxyxanthone which is found in the seeds of Mammea americana:
540
The Synthesis of Oxygen Ring Compounds
The intramolecular acylation that concludes the synthesis is much more easily accomplished than its intermolecular counterpart and the weakest part of the sequence lies in the diphenyl ether formation. There can also be problems of orientation in the cyclization step; Goldberg and Wraggls7 have discusscd some elementary situations in mechanistic terms. The Kielmeyeru xanthones 227 and 228 cannot be prepared by the GSS method because of dealkylation but Dallacker and Dam6 offer the following solution'** t o the preparation of the former; the use of activated compounds to improve the Ullmann synthesis is noteworthy, as is the mildness of the cyclization conditions:
H#'d;
I
HNO? Cu10
AcCI, H 2 S 0 4 a1 60"
The phenol 228 was made similarly, with the benzyloxy group instead of methoxy at the critical stages. However, the greatest successes of the route have been in the sterigmatocystin area (discussed earlier) where it was absolutely essential to employ none but the mildest reagents.
7. Xanthones
541
Recent innovations have again hinged upon benzophenone intermediates and. their cyclization. Stout, Balkenhol, and their colleagueslsB~lBO combine methyl ethers (rather than phenols) with acid chlorides in true FriedelCrafts reactions and cyclize the resulting benzophenone in basic conditions (tetramethylammonium hydroxide in pyridine). They have prepared a number of the highly rnethoxylated constituents of species of Gentianaceae in this manner; two examples will serve: (a) a synthesis of I-hydroxy-2,3,4,7tetramethoxyxanthone 229, which is found in the root of Frasera caroliniensis; and (6) a synthesis of I ,2,3,7-tetraniethoxyxanthone230:
1
Me0
M
McO \
HO 0
229
e
Me0\ Me0
o 0
\ mOMe 230
Both examples include selective dernethylation by aluminum chloride of compounds with many methoxy groups at risk in so acid a medium. The gentlest possible technique is used (though boron chloride might serve the purpose better), and the only methoxy group affected is one subject to both coordination and to buttressing (or to a second incursion of the coordination
542
The Synthesis of Oxygen Ring Compounds
phenomenon). We can now complete a series of flanking groups rendering methoxyl groups labile; in order of increasing efficiency this runs :one methoxy group, one carbonyl group, two methoxy groups, one methoxy with one carbonyl group, two carbonyl groups. If desired, complete demethylation can be accomplished and the ring closed giving a polyhydroxyxanthone in one step. This is conveniently done using pyridinium chloride in a fashion prescribed by French authors.lQ1 The most interesting feature is the base-catalyzed cyclization. The recognition of the possibility is due to Barton and Scott,lB1who pointed out that the benzophenone carbonyl group should promote nucleophilic additionelimination reactions :
Considerable advances have been made following suggestions by LewislS2 that many naturally occurring xanthones might arise by oxidative coupling in benzophenone precursors and his demonstration of the following reaction in vitro
Several schools have elaborated upon the idea,1Q2-1eswhich receives impetus from tracer showing that benzophenones can indeed act as xanthone precursors and from the (occasional) cooccurrence of both the
7. Xanthones
543
xanthones that are theoretically possible when the cyclization can take alternative direction^.'^^ Chemical oxidation is usually reported as giving only the product of para coupling, but Atkinson and Lewis196have succeeded in isolating small amounts of the products of ortho coupling, as in the following instance in which two xanthones, 231 and 232,found together in Mammea arnericana are formed simultaneously by ferricyanide oxidation :
53% 1.1
232
Oxidative cyclization of 2,3',6-trihydroxybenzophenone gave a similar pair of xanthones, permanganate proving the better oxidant in this case:
0
On the other hand, oxidation of the tetrahydroxybenzophenone233 resulted only in the para coupled product, the plant xanthone gentisein 234, a pigment of Gentiuna lutea. In separate experiments it was demonstrated that the ortho coupled product 235 is so easily oxidized further that it would have been destroyed before detection.lg5This marks a rather serious limitation of the method, though it may perhaps be mitigated by the use of partly methylated
544
The Synthesis of Oxygen Ring Compounds
xanthone 235 has not been recognized as such in plants, although its ethers have been.
benzophenones-interestingly,
0
233
0
235
These oxidations are also known to be dependent not only on time of exposure to the reagent but also on the pH of the solution. Moreover, the enzymes peroxidase and laccase are capable of bringing about oxidative cyclization in vifro, ortho, and para coupling being observable.1Q5Such cyclizations are generally held to have radical mechanisms demanding a phenolic hydroxyl group in the ring suffering substitution, a possible exception being DDQ oxidations which succeed with para methoxy groups suggesting electrophilic substitution and phenoxonium (phenoxylium) ion intermediates. 198
Somewhat similar studies by Ellis, Whalley, and disclose a different mechanism, oxidation to a quinone intervening. For example, ferricyanide oxidation converts 236 into the trihydroxyxanthone (237),presumably by a radical coupling mechanism, whereas some other oxidants supply the quinone 238, which rapidly cyclizes in warm methanol to give the same product. Another interesting variation has been provided by Jefferson and S ~ h e i n m a n nwho , ~ ~ discovered ~ that the benzophenone maclurin 239 is best oxidised to the xanthone 240 (in 45% yield) by photochemical means. Although the product 241 of ortho coupling was not detected (it may have been destroyed concurrently), all three compounds do occur together in the plant Symphonia g l ~ b u l i f e r a .Similar ~ ~ ~ photochemical cyclizations have
mqp 75
x
545
0tl
0
236
7. Xanthones
0
iDDQ 0
100%
II
237
T
238
given 1,6,7-trihydroxyxanthone, a constituent of Garcinia eugenfolia and Manirnea africana. Unfortunately, the method has failed entirely in other cases even when ordinary chemical methods have been successful, and at present there seems to be no easy way of accounting for the differences.200 Ho&lOpOH
+ hv
OH 0
239 Maclurin
\
241
'i, ""O \
T
O\
0
H
240
OH
The benzophenones required for these studies can be made by standard techniques; that evolved by Usgaonkar and Jadhavzolhas proved particularly useful. For various reasons, however, some of which have already been touched on, derivatives of 2,6-dihydroxybenzophenonetend to be troublesome to prepare, yet, because they happen to be important, Locksley and MurrayZoohave devised a general route to them (compare 46). An example follows overleaf:
546
O
The Synthesis of Oxygen Ring Compounds
M e Li
+
M L‘o COCl
__f
46 %
v:: OMc
0
OMe
45 %
We can note three other methods briefly. That introduced by TanaseZo2 utilizes xanthylium salt intermediates. Paquette’s methodzo3takas advantage of the synthetic powers of enamines, and Guyot and Mentzer’s methodzo4 of the simplicity of noncatalyzed thermal condensations between suitably protected phenols and ethyl cyclohexanone-2-carboxylate.These last two methods require dehydrogenations for their completion. We now leave the synthesis of the simpler xanthones to consider ways of building them up into more complex compounds. Scandinavian workerszo5 find that chlorine in acetic acid converts norlichexanthone 208 into thiophanic acid 242, a constituent of the lichen Lecanora rupicola. With a limited amount of chlorine the resorcinol ring is less attacked than the phloroglucinol ring (as would be predicted) so that some arthothelin 243 can be obtained. This is found in another lichen, L. straminea, and with diazomethane rapidly gives thuringion 244. The selectivity of this reaction depends upon the very powerful acidifying effect that two ortho chlorine atoms have on a phenolic hydroxyl group. Finally, Raney nickel dehalogenation gave griseoxanthone C 245, the sequence representing a most unusual way of effecting selective methylation in a xanthone derivative:
0
242 Thiophanic acid
0
243 Arthothelin 244 Thuringion
R =H R = Me
245 Griseoxanthone C
0
7. Xanthones
547
Belladifolin 246 occurs in Gentiana belladifolia. To synthesize it, Markhamzo6 neatly employed persulfate oxidation and selective methylation beginning with 1,3,8-trihydroxyxanthonc.Neither reaction is truly selective, however, for oxidative attack on the phloroglucinol ring probably occurs and the methylation depends upon the enhanced acidity of hydroxyl para to a carbonyl group, an effect that spectroscopic s t ~ d i e s ' 7 ~show - ' ~ ~ to be much less clear cut in the xanthone than in other series.
wol' OH
+
\
0
K,S,O,
OH
1-
%OH
OH
OH
n v
246 Belladifolin
The synthesis of polygalaxanthone-B 248 by Jain, Khanna, and Seshadri20' began with xanthone 247 and illustrates a useful combination of Duff and Dakin reactions for providing hydroxyl groups often met with among other pyrones, especially the coumarins
-1
247
DMSiKtC03
woMc OMc
Me0 \
\
0
OH
OH
DMS
__f
W
McO\
248
O
OMc M
c
OMc OMe
\
548 The Synthesis of Oxygen Ring Compounds
Formyl groups introduced by the Duff reaction or acetyl groups introduced by some form of Friedel-Crafts reaction have been used to build up furanoxanthones and other more complex combinations of rings despite the absence of any examples known to occur n a t ~ r a l l y . We ~ ~ meet ~ ~ ~in~ 0the example one of the most important orientational specificities among hydroxyxanthones:
Specific 2-substitution characterizes most nuclear alkylation reactions of 1,3-dihydroxyxanthone derivatives. An important example is the synthesis of one of the more common xanthones, mangiferin 249, with its unusual C-glucosyl substituent. This is introduced by treating I ,3,6,7-tetrahydroxytetraacetate in methanolic xanthone with I-bromo-a-D-g~ucopyranosyl sodium methoxide, with or without sodium iodide; not surprisingly, the yield is very poor.211s212
A Mangiferin 249
Another example of this orientational effect occurs in routes to osajaxanthone 250; as with the acetylxanthone mentioned previously, the orientation is observed whether the substituent is introduced into a preformed xanthone nucleusa13or the xanthone ring is formed last.17a,17s-214
7. Xanthones
0
549
1"+
Osajaxanthone 250
For the first fully definitive synthesis of apyranoxanthone, jacareubin 255, a pigment of the heartwoods of Calophyllum brasiliense Camb. and C. sclerophyllum Vesq., Jefferson and Scheinmann21sbegan with 251, having introduced the C, substituent by means of a combination of dimethylallyl bromide and silver oxide, which does not work well but is much better than the more usual bromide-sodium methoxide mixture (perhaps ion pairs are involved) :
550
The Synthesis of Oxygen Ring Compounds
Next, hydriodic acid deetherified the compound and cyclized it, both isomers (252,253) being obtained, although chelation of the 1-hydroxy group would be expected to oppose the formation of one. It may be that in the very strongly acidic solution the nucleus is protonated so that cyclization actually occurs in the salt 254 where chelation would be unimportant: HO 251 HI , HO
,
'
OH
252
0
HO 253
254
(rn(rn3 [rn-
In agreement, use of a weaker acid (formic acid) is known to favor the linear isomer in an equivalent case.213 The final stages in the synthesis of jacareubin 255 can be summarized thus :
fACtO
252
H \b & p y 0
OH
Jaca reubin 255
HO-
AcO
Br,
AcO
Br
7s P
w
0
c
AcO
AcO
Br
7. Xanthones
551
The foregoing scheme includes bromination-dehydrobromination as a method of converting a chroman into a chromene, a practice going out of fashion now that gentler and more easily controlled syntheses are available; the dibromination sequence came to light during parallel work on deoxyjacare~bin'~'(starred groups omitted). A direct synthesis of jacareubin 255 by Crombie's method was mentioned earlier but without reference to the orientational problem. This can be taken up now, routes to deoxyjacareubin offering the opportunity. One approach to deoxyjacareubin 257 takes advantage of the Claisen rearrangcment of the simple ally1 ether 258, which readily affords 259, protection of the phenolic group and oxidative fission of the olefinic link following.177The resulting aldehyde 260 has many synthetic uses; sequences of this kind were originally introduced for the synthesis of furopyronesz1" and have recently been employed174in elaborating the side chain of scriblitifolic acid 266. The present example"' continues with the appropriate Wittig reaction needed to complete the desired dimethylallyl substituent, and after that boron chloride serves to remove one methyl group giving the phenol 262.
257 Deoxyjacareubin R = H
MWe 0
O
M
e
CHzCHO OMe
\
\
-
M~e 0
O
\
\
0
M
H
C
CHpCH=CH,
259
M eO
261
0 263
R
=H
A second methyl group is removed by hot aqueous piperidine yielding xanthone 263. This reagent preferentially attacks methoxy groups para to carbony1 groups, and the preference may be understood in terms of the vinylogous
552
T h e Synthesis of Oxygen Ring Coiiipounds
ester system comprised by that arrangement. An orflzo carbonyl group could serve similarly but, as is usually the case, orfho activation is less than para activation. The formation of buchanaxanth~ne~’~ 265 from xanthone 264 is another example from the recent literature, but it must be noted that the degree of selectivity seems to vary considerably and unpredictably.
0
0 Scriblitifolic acid 266
Buchanaxanthone
264 R = M e 265 R = H
The synthesis is completed with DDQ oxidation of 263 to give the desired linear product, deoxyjacareubin 5-methyl ether (257; R = Me), since the other hydroxyl group is protected by chelation. This result definitely orients the ring fusions in deoxyjacareubin, but this compound has not been prepared from the ether. Demethylation of the last methoxy group would not be easy and would endanger the pyran system. Deoxyjacareubin itself has been ~ynthesized’~’ by methods used for jacareubin. Application of the Crombie chromene synthesis to the deoxyjacareubin problem discloses a curious anomaly. As Lewis and Reary have shown, the condensation of 1,3,5-trihydroxyxanthonewith 3-methylcrotonaldehyde in boiling pyridine yields deoxyjacareubin 257 directly.218 Other workers, however, find that I ,3-dihydroxy-5-methoxyxanthone267 yields only the “wrong” isomer 268 in dilute solution, thus abrogating the orientation rule strictly obeyed in all the previous discussion. The secret lies in the concentration; for when this is high, the linear compound 269 does make an appearance.”’ One might guess that, since the boiling point of the concentrated
268
f
261
(ypy \
\
0
OH
269
7. Xanthones
553
solution must be considerably higher than of the dilute solution, we may be witnessing here a changeover from kinetic to thermodynamic control. The appropriate studies have yet to be made. Alvaxanthone 270 possesses a different type of prenyl group. An obvious synthetical plan would be based on a Claisen rearrangement in the ether 271; this works, but not well.180Thedifficulty lies in fitting the bulky branched end of the C , group between the hydroxy and methoxy groups, and it has two consequences. The first is that the migrating group continues on its way, with another inversion, to the para position; and, since there is now no marked steric effect, 272 is the chief product. The second is that ring closure of the desired compound 273 is promoted, so the small yield is depleted further giving the furanoxanthone derivative 274,
Alvaxanthone
270
& OH
272
Another fact that has to be kept in mind is that in this series prenyl ethers may have an unexpected sensitivity to silica.21eAttempts to purify 275 on silica columns lead to complex mixtures containing fair quantities of the 2-alkenylxanthone 276 not obtainable by Claisen rearrangement. It is
554
The Synthesis of Oxygen Ring Compounds
believed that silica is acting as an acid* and inducing ion pair formation as in 277 (cf. p. 549). Whatever the explanation, the chromatographic analysis of natural materials has to be conducted with particular caution where allylic materials might be involved. Other workers have also noted the acidity of silica; xanthones sometimes show marked color changes on this column packing; and, because phenols are less basic than their ethers in the xanthone series, inethylation induces a fall in R, instead of the usual rise.180 McO W \
O
\M
0
Me0W \
e SlOl --f
O 0 2”
275
\M
0-
e +
C H,C H =CMe,
I
1
In conclusion, we consider the most novel recent advance in the chemistry of prenylxanthones. At the beginning of this chapter it was pointed out that some Claisen rearrangements might lead to bridged xanthone nuclei of the kind that characterize gambogic acid , the morellins, and bronianone. During the writing, the idea has been brought nearer actuality by Quillinan and ScheinmannlZzowho have demonstrated that the 5,G-diallyl ether 278 of jacareubin gives just such a bridged system 279 when heated in boiling decalin for 14 hours. (It should be noted that in 6-allyloxyxanthones migration is normally to position 5 as shown in the first stage.) The second stage is merely an internal Diels-Alder addition. Despite the severity of the reaction conditions in this instance, the reaction may well have biosynthetical implications. A final note on spectroscopy. Xanthones have changed in the last few years from little known compounds to well studied ones, and considerable spectroscopical information is now available that helps enormously in deciding structures and particularly orientations in synthetical sequences. Quantities of
* In work not yet published21e it was found that only prenyl ethers at position 1 are thus affected. This is entirely in accord with acid catalysis.
References
555
'I
0 279
REFERENCES W. D. Ollis and 1. 0. Sutherland, in W. D. Ollis, Ed., The Chemistry ofNatural Pltenolic Materials, Pergamon, Oxford, 1961, Chapter 4. 2. F. M. Dean, NufurallyOccurring Oxygen Ring Compounds, Butterworths, London, 1963, Chapter 7. 3. W. B. Whalley, in W. D. Ollis, Ed., The Chemisfry ofNafural Phenolic Materialq Pergarnon, Oxford, 1961, Chapter 2. 4. W. D. Ollis. M. V. J. Rarnsey, I. 0. Sutherland, and S . Mongkolsuk, Tetrahedron, 1.
21, 1453 (1965).
5. 6.
7. 8.
R. Livingstone and R. B. Watson, J. Chem. Soc., 1956, 3701. L. I. Smith and P. M. Ruoff, J . Am. Chem. SOC.,62, 145 (1940). J. N. Chatterjea, J. Indian Chem. Soc., 36, 76 (1959). J. Cottam, R. Livingstone, M. Walshaw, K. D. Bartle, and D. W. Jones, J. Chent. Soc., 1965, 5261 C. S. Barnes, M. I. Strong, and J. L. Occolowitz, Tefmhedrot~.19, 839 (1963). T. R. Kastuti and T. Manithomas, Tetrahedron Lett., 1967, 2573. E. Spath and R. Hillel, Ber. dtsch. chem. Ges., 72, 963, 2093 (1939); E. Spath and H. Schmid, Ber. drsclt. chem. Ges., 74, 193 (1941). G. Cardillo and L. Merlini, Gazretfa,98, 191 (1968). S. K. Mukerjee. S. C. Sarkar, and T . R. Seshadri, Tetrahedron, 25, 1063 (1969). I
9. 10.
11. 12. 13. '
556
The Synthesis of Oxygen Ring Compounds
14. J. Nick], C/iettt. B p i - . , 91, 1372 (1958). H. D. Schroeder. W . Bencze, 0. Halpern, and H. Schmid, Cherft. Ber., 92, 2338 (1 959). 16. I. lwade and J. lde, Clietri. Pharm. Bid/. (Japot1), 10,926 (1962); 11, 1042 (1963). 17. J. Hlubucek, E. Ritchie, and W. C. Taylor, Tetrahedron Lett., 1969, 1369. 18. J. Zsindely and H. Schmid, Helu. Chitti. Acta, 51, 1510 (1968). 19. R. D. H. Murray, M. M. Ballantyne, and K. P. Mathai, Tetrahedrorr Lett., 1970, 243. 20. A. J. Birch and M. Maung, Tefroltedrori Lett., 1967, 3275. 21. L. Edwards, J. Kolc, and R. S . Becker, Plioroclierti. Photobiol. 13, 423 (1971); J. Kolc and R. S. Becker, J. Pliys. C/teni., 71,4045 (1967). 22. A. B. Turner, Q. Xeu., 18, 347 (1964). 23. I. M. Campbell, C. H. Calzadilla, and N. J. McCorkindale, Terruhedrotr Lett., 1966, 5107. 24. E. A. Brande, L. M. Jackman, R. P. Linstead, and G. Lowe, J. Cltetn. Soc., 1960, 3133, 3144. 25. R. Mechoularn, B. Yagnitinsky, and Y. Gaoni, J. Atti. Cltetn. Soc., 90,2418 (1968). 26. G . Cardillo, R. Cricchio, and L. Merlini, Tefroltedroti, 24,4825 (1968). 27. J. A . Miller and H.C. S. Wood, C/iet~t.Cuttitti., 39, 40 (1965). 28. L.Jurd, K. Stevens, and G. Manners, Tetr.a/iedroti Lett., 1971, 2275. 29. F. Bohlmann and K. M. Kleine, Clierti. Ber., 99, 885 (1966). 30. A. C. Jain and M. K. Zutshi, Terraliedroti Left., 1971, 3179. 31. D. E. Games and N. J. Haskins, Clieni. Conitti., 1971, 1005. 32. J. R. Beck, R. Kwok, R. N. Booker, A. C. Brown, L. E. Patterson, P. Pranc, B. Rockey, and A. Pohland, J. Atti. Clietn. Soc., 90.4706 (1968). 33. L. Merlini and R. Mondelli, Tetrahedroti, 24, 497 (1968). 34. J. Links, Eioc/tetu. Biophp. Arta, 38, 193 (1960). 35. F. W. Hemming, R . A. Morton, and J. F. Pennock, Bioche/tr.J., 80, 445 (1961). 36. D. McHale and J. Green, Chetti. Ittd., 1962, 1867; J. Clierri. Soc., 1965, 5060. 37. B. 0. Linn, C. H. Shunk, E. L. Wong, and K. Folkers, J. Atti. Chetti. Soc., 85,239 (1963). 38. H. W. Moore and K. Folkers, Atiti., 684, 212 (1965). 39. R. Huls, Bull. Soc. Cliirn. Be&., 67, 22 (1958). 40. F. Baranton, G. Fontaine, and P. Maitte, Btill. Soc. Chiiti. Frotire, 1968,4203. 41. A. K. Ganguly, B. S . Joshi, V. N. Kamat, and A. H. Manmade, Tetrulrcdroti, 23,4777 (1967). 41a. J. R . Collier and A. S. Porter, Cltern. Conirn., 1972, 618. 42. W. Bencze, J. Eisenbeiss, and H. Schmid, Helu. Chitti. Acru, 39, 923 (1956). 43. T. R. Seshadri and M. S. Sood, Tetrahedron Lett., 1964, 3367. 44. S. N. Shanbag. Tetrahedroti, 20, 2605 (1964). 45. K. Hata and K. Sano. Tetralterfrori Lett., 1964, 3367. 46. S. Nozoe and K. Hirai, Tetruhedrori Lett., 1969, 3017. 47. R. M. Bowman, M. F. Grundon, and K . J. James, Chenr. Cornrti., 1970,666. 15.
References 48. 49.
557
L. Crombie and R. Ponsford, Chern. Comm., (a) 1968, 368; (b) 1968, 894. V. Kane and R. K.Razdan, J. Am. Client. Soc., 90,6551 (1968); Tetraliedrotr Lett.,
1969, 591. 50. M. J. Begley, D. G. Clarke, L. Crombie. and D. A. Whiting, Chern. Cotirnr.. 1970, 1547. 51. L. Crombie, R. Ponsford, A. Shani, B. Yagnitinsky. and R. Mechoulani, TerraIiedron Len., 1968, 5771. 52. S . P. Kureel, R. S. Kapil, and S. P. Popli, Cheni. hid., 1970, 958; Chenr. con it^, 1969,1120. 53. W. M. Bandaranyake, L. Crombie, and D. A. Whiting, Cheni. Conrnr., (a) 1969,58; (b) 1959, 970. 54. W. J. G. Donnelly and P. V. R. Shannon, Cheni. Cottitti., 1971, 76. 55a. E. E. Schweizer, J. Liehr, and D. J. Monaco, J. Org. Cheni., 33, 2416 (1968);
55b. 56.
57. 58. 59. 60.
E. E. Schweizer, E. T.Schaffer, C. T. Hughes, and C. J. Berninger, J. Org. Cliem., 31, 2907 (1966). N. S. Narazimhan, M. V. Paradkar, and A. M. Gokhale, Tetrahedron Lett., 1970, 1664. K. Sargeant, R. B. A. Carnaghan. and R. Allcroft, Client. hid., 1963, 53; K. J. van der Merwe, L. Fourie, and de B. Scott, Chenr. bid., 1963, 1660; F. Dickens and H. E. H. Jones, Br. J. Cancer, 19, 392 (1965); G. N. Wogan, Bacferiol Reu., 30, 460 (1966); P. C . Spensley, Edeauotw, 22, 75 (1963); C. W. Hesseltine, 0. L. Shotwell, J. J. Ellis, and R. D. Stubblefield, Backviol. Reu., 30, 795 (1966). C. W. Holzapfel; P. S. Steyn, and I. F. H. Purchase, Tetrahedrorr Left., 1966,2799; 1. F. H. Purchase, Fed. Costnet. Towicol., 5, 339 (1967). E. Bullock, J. C. Roberts, and J. G. Underwood, J. Clrerrr. Sor., 1962, 4179. T. Haniasaki, M. Renbutsu, and Y. Hatsuda. Agric. B i d . Clreitr. (Japan), 31, 11 (1967). C. A. Bear, J. M. Waters, T. N. Waters, R. T. Gallagher, and R. Hodges, Client. Coi~im.,1970, 1705.
61. T. Asao, G. Buchi, M. M. Abdel-Kader, S . B. Chang, E. L. Wick, and G. N. Wogan, J. Anr. Cheiir. Soc., 87, 882 (1965). 62. H. E. Zaugg, F. E. Chadde, and R. J. Michaels, J . Anr. Chiti. Soc., 84,4567 (1962). 63. E. C. M. Coxworth, Can. J. Clrenr., 45, 1777 (1967); B. S. Thyagarajan, K. K. Balasubramanian, and R. Bhima Rao, Can. J. Client., 44, 633 (1966). 64. C. H. Eugster and P. Karrer, Clrirnia, 18, 358 (1964). 64a. J. S . E. Holker, personal communication. 65. P. L. Sawhney and T. R. Seshadri, Pror. bidian Arad. Sci., 37A, 592 (1953). 66. G . Buchi, D. M. Foulkes, M. Kurono, G. F. Mitchell, and R. S. Schneider,J. h i . Chem. Soc., 89, 6745 (1967). 67. J. A. Knight, J. C. Roberts, and P. Rofey, J . Clieiir. Sor., 1966, 1308. 68. A. Schiavello and E. Cingolani, Garzetta, 18, 717 (1951). 69. J. A. Knight. J. C . Roberts, P. Roffey, and A. H. Sheppard, Cheni. Cor~rnr.,1966, 706. 70. J. C. Roberts, A. 1968,22.
H.Sheppard, J. A. Knight, and P. Roffey, J . Chett~.Soc. (C),
558
The Synthesis of Oxygen Ring Compounds
S. Brechbuhler, G.Biichi, and G. Milne, J . Org. Client., 32, 2641 (1967). J. G.Underwood and J. S. E. Holker, Cheni. bid., 1964,1865. C.L. Lange, H. WanhofT, and F. Korte, Chein. Ber., 100,2312 (1967). G.Buchi and S. M. Weinreb, J. Am. Client. Soc., 69,5408 (1969). S. Sethna and R. Phadke, Org. Reactions, 7, 12 (1953). B.W.Bycroft, J. R. Hutton, and J. C. Roberts,J. Cheni. Soc. (C),1970,281. M. J. Rance and J. C. Roberts. Tetrahedron Leu., 1969,277;1970,2799. R. J. Cole and J. W. Kirksey, Tetrahedroti Lett.. 1970, 3109 G.C. Elsworthy, J. S. E. Holker, J. M. McKeown, J. B. Robinson, and L. J. Mulheirn. Chem. Comm., 1910, 1069; M. Biollaz, G. Biichi, and G. Milne, J. Ant. Chein. SOC.,92, 1035 (1970). 80. E.Bullock, D. Kirkaldy. J. C. Roberts, and J. G. Underwood, J. Client. Soc., 1963, 829;G . M. Holmwood and J. C. Roberts, Tetrahedron Lerr., 1971, 833. 81. J. S. Davies, V. H. Davies, and C. H. Hassall, J. Chern. SOC.(C),1969, 1873; C. H. Hassall and B. A. Morgan, Cheitt. Contin., 1970, 1345. 82. L. Crombie and R. Peace, Chem. Conim., 1963, 246. 83. G.Buchi, L. Crombie et al., J . Chew. Soc., 1961,2861. 84. F.B. LaForge, H. L. Haller, and L. E. Smith, Chein. Reu., 12, 182 (1933); L. Feinstein and M. Jacobson, Fortschr. Cheni. Org. Nuiurst., 10, 436 (1953);H. L. Haller, L. D. Goodhue, and H. A. Jones, Cheni. Rev., 30, 33 (1942). 85. N. Nakatani and M. Matsui, Agric. Biol. Chetn. (Jupan), 32, 769 (1968). 86. D.J. Ringshaw and H. J. Smith, J . Cheitr. Soc. (C),1968; 102. 87. F. B. LaForge and H. L. Haller, J . Am. Chenr. Soc., 54, 810 (1932). 88. H.Fukami, M. Nakayama, and M. Nakajima, Agric. Biol. Chem., 25, 243 (1961). 89. 0.Dann and G . Volz, Jusriis. Liebigs Atin., 631, 102, 1 1 1 (1960). 90. G.Buchi and L. Crornbie, J. Chetn. Soc., 1961,2843. 91. M. Miyanoand M. Matsui, Client. Ber., 92, 1438,2487 (1959);&ill. Agric. Cheni. Soc. Japan, 24, 540 (1960). 92. L. Crombie and J. W.Lown, Proc. Chein. SOC.(London), 1961,299. 93. M. Miyano, J. Am. C/tetn. Soc., 87, 3962 (1965). 94. A. Robertson and G. Rusby, J. Chetn. Soc., 1936, 212. 95. J. R. Herbert, W. D. Ollis, and R. C. Russell, Proc. Chein. Soc., 1960, 177. 96. M. Miyano, J. Ain. Cheni. Soc., 87,3958 (1965). 97. M. Baran-Marszak, J. Massicot, and C. Mentzer, Cottip/. rend., 263C, 173 (1966).
71. 72. 73. 74. 75. 76. 77. 78. 79.
98. 99. 100. 101.
T.R. Seshadri and S. Varadarajan, Proc. Indian Acad. Sci., 37A, 784 (1953). A. C.Mehta and T. R. Seshadri, Proc. Iiidian Acud. Sci., 42A, 192 (1955). Y.Kawase and C. Nurnata, Chent, Ind.. 1961, 1361.
V. Chandrashekar, M. Krishnamurti, and T. R. Seshadri, Terraliedrort, 23, 2505
(1967). 102. K.Aghoramurthy, A. S. Kubla, and T. R. Seshadri, J . Itidinn Clietn. SOC.,38,914 (1 961). 103. K.Fukui, M. Nakayarna, and T. Harano. Bull. Chein. SOC.Jupaii, 42, 1693 (1969). 104. G. A. Caplin, W. D. Ollis, and I. 0. Sutherland, J. Chern. SOC.(C), 1968,2302.
References 105. 106. 107. 108.
L. Cronibie, P. W. Freeman, and D. A. Whiting, Cheni. Conirn., 1970,563. L. Crombie, C. L. Green, and D. A. Whiting, J. Cheni. SOC.( C ) , 1968,3029. V. R. Sathe and K. Venkataraman, Ciirrent Sci., 18,373 (1949).
559
W. D. Ollis. K. L. Ormand. and I. 0. Sutherland, J. Chem. SOC.(C), 1970. 119; W. D. Ollis, K. L. Ormand, B. T. Redman, R. J. Roberts, and I. 0. Sutherland, J . Client. Soc. (C), 1970, 125. 109. C. A. Weber-Schilling and H. W. Wanslick, Terraliedrori Lett., 1964,2345. 110. E. K. Adesogan, F. M. Dean, and M. L. Robinson, unpublished observations. 111. W. J. Bowyer, A. Robertson, and W. B. Whalley, J. Cliem. Soc., 1957, 542. 112. N. R. Krishnaswamy and T. R. Seshadri, J. Sci. Ind. Res. India, 16B, 268 (1957). 113. A. H. Gilbert, A. McGookin, and A. Robertson, J. Chem. Soc., 1957, 3740. 113a. T.R. Govindachari, K. Nagarajan, and P. C. Parthasarathy, J. Client. Soc., 1957, 548; 0.H. Emerson and E. M. Bickoff, J. Am. Clretn. Soc., 80, 4381 (1958); Y . Kawase, Biill. Cliem. Soc., Japan, 32, 690 (1959). 113b. C. Deschamps-Vallet and C. Mentzer, C o t p t . Rerrdii., 251, 736 (1960). 114. V. K. Kalra, A. S. Kubla, and T. R. Seshadri, Terrohedron Lett.. 1967,2153. 115. D. Nasipuri and G. Pyne, J. Sci. Indust. Res. India,21B, 51 (1962). 116. K. Fukui and M. Nakayama, J. Clieni. SOC.Jupun, 85, 790 (1964). 117. H.-W. Wanslick, R. Gritsky, and H.Hcidepricm, Cheni. Ber., 96, 305 (1963). 118. M. Darbarwar, V. Sundararnurthy, and N. V. Subba Hao, Cltrrent Sci., 38, 13 (1 969). 118a. H. W. Wanslick, in W. Foerst, Ed., Newer Methods of Preparitive Organic Chemistry, Academic Press, New York, 1968,p. 139. 119. D. Bouwer, G. v. d. M. Brink, J. P. Engelbrecht, and G. J. H. Rall, J. S. African Clietri. Inst.,21, 159 (1968). 120. K. Fukui and M. Nakayama, Tetrahedron Lert., 16, 1805 (1966); 1965,2559. 121. L. L. Sinionova and A. A. Shamshurin, Zli. Vsesoyriz. Khitti. Obslich. h i . D. I. Il.lertde/eroa, 11, 252 (1966); Clicrii. A h . , 65, 10572 (1966);Clieiii. A h . , 67, 64380 (1967). 122. K. Fukui, M.Nakayania, and H. Sesita, But/. Clreni. Soc., Japan, 37, 1887 (1964). 123. W. Dillhey and W. Hoschcn, J . Prakt. Cheni,, 246, 42 (1933); P. Karrer, R. Widmer, A. Helfenstein, W. Hurliman, 0.Nievergelt, and P. Monsarrat-Thomas, Helv. Cltini. Actn. 10, 729 (1927). 124. L.Jurd, Tetrahedron Lc/t., 1963, 1151;J . Org. Clietn., 29, 2602 (1964). 125. L.Jurd, Te/raliedron, 1966,2913. 126. L. Jurd, J . Org. Client,, 1964,3036. 127. A. L. Livingstone, S. C. Witt, R. F. Lundin, and E. M. Bickoff, J . Org. Cltenr., 30,2353 (1965). 128. L.Jurd, J . Pharni. Sci., 54, 1221 (1965). 129. R. R. Spencer, B. E. Knuckles, and E. M. Bickoff, J. Heterocycl. Chew., 3, 450 (1966). 130. K. Fukui, M. Nakayania, H. Tsuge, and K. Tsuzuki, Experientiu, 24, 536 (1968). 131. A. L.Livingstone, E. M. Bickoff. R. E. Lundiii, and L. Jurd, Tetrahedron, 20,1963 ( I 964).
560 132. 133. 134. 135. 136. 137. 138. 1 39. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 152.
153. 154. 155. 156. 157. 158.
159. 160. 161. 162. 163. 164. 165.
The Synthesis of Oxygen Ring Compounds R. R. Spencer, B. E. Knuckles, and E. M. Bickoff, J. Org. Clteni., 31,988 (1966). Jurd, J. Org. Clienr., 27, 872 (1962). P. M. Dewick, W. Barz, and H. Grisebach, Client. Corrtni., 1969, 466. G. W. K. Cavill, F. M. Dcan, J. F. E. Kcenan, A . McGookin, A. Robertson, and G. B. Smith, J . Cliem. SOC., 1958, 1544. S. H. Harper, A . D. Kemp, and W. G.Underwood, Clterii. Corifn?.,1965, 309. K. Fukui, M . Nakayarna, and T. Harano, Bull. Cliem. Sor. Japari, 42, 233 (1969). K. Fukui, M. Nakayama, and K. Tsuzubi, Experku/ia, 25, 122 (1969); Bull. Cltetii. Sor. Japan, 42,2395 (1969). K. M. Biswas, L. E. Houghton, and A. H. Jackson, Tetrahedrofi Suppl., 22(7), 261 (1966). H. Suginome and T. Iwadare, Bull. Chefti. Soc., Japan, 33, 567 (1960). M. Uchiyama and K. Ooba, J. Agric. Client. SOC. Jnpart, 42, 688 (1968). K. Fukui and M. Nakayama, Bit//. Clrerrt. SOC. Japan, 42, 1408 (1969). V. K. Kalra, A. S. Kubla, and T. R. Seshadri, bidan J . Client., 4, 201 (1966). K. Fukui, M. Nakayama, and T. Harano, B d . Clierti. SOC., Japan, 42, 1693 (1969). M. Uchiyama and M. Matsui, Agrir. Biol. Client., 31, 1490 (1967). C. W. L. Bevan, A. J. Birch, B. Moore, and S. K. Mukerjee, J . Clietrt. Soc., 1964, 5991. W. B. Whalley, Pitre Appl. Cliem., 7, 565 (1963). F. M. Dean, J. Staunton, and W. B. Whalley, J . Clietii. Soc., 1959,3004. S . Yarnamura, K. Kato, and Y.Hirata, Te/raliedrofi Lelt., 1967, 1637. L. L. Woods, J . Anr. Clietri, SOC.,80, 1440 (1958); M. Ohta and H. Kato, B d . Clietii. SOC. Japan, 32, 107 (1959). G.R. Birchall, M. N. Galbraith, and W. B. Whalley, Clteni. Co/ttrn., 1966, 474. F. Wessely, G. Lauterbach-Keil, and F. Sinwel, Moria/slt., 81, 811 (1950); F. Wessely, E. Zbiral, and H. Sturm, Chefit. Eer., 93, 2840 (1960). M. N. Galbraith and W. B. Whalley, Client. Comniirri., 1966, 620. G. R. Birchall, R. W. Gray, R. R. King, and W. B. Whalley, Client. Comti., 1969, 457. R. Chong, R. R. King, and W. B. Whalley, Chem Cmiiui., 1969, 1512. Reference 2, Chapter 15. E. J. Haws, J. S. E. Holker, A. Kelly, A. D. G. Powell, and A. Robertson, J. Client. Soc., 1959, 3598. B. C. Fielding E. J. Haws, J. S. E. Holker, A. D. G. Powell, and A. Robertson, Tetraliedrorr Lett., 1960, 24. R. Chong, R. W. Gray, R. R. King, and W. B. Whalley, Chem Comrir., 1970,101. G . Buchi, J. White, and G. N. Wogan, J . Ani. Client. Soc., 87, 3484 (1965). W. E. Truce, Org. Reactions, 9, 38. D. H. Johnson, A. Robertson, and W. B. Whalley, J. Client. Sor.. 1949, 1563. H. H. Warren, G. Dougherty, and S. E. Wallis,J. Am. Clrern. SOC., 79,3812 (1957). R. F. Curtiss, C. H. Hassall, and M.Nazar, J. Clrern. SOC. (C), 1968,85. D. H. R. Barton and J. B. Hendrickson, J. Clieni. SOC., 1956, 1028.
. L.
References 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 178. 178a. 179. 180. 181. 182. 183. 184. 185. 186. 187. 188. 189. 190. 191. 192. 193. 194. 195. 196. 197. 198. 199.
561
0.R. Gottlieb, Phytochern., 7,411 (1968). I. Carpenter, H.D. Locksley, and F. Scheinrnann, P/iy/ocliern.,8, 2013 (1969). J. C.Roberts, Cliew. Rev., 61, 591 (1961). Reference 2, Chapter 9. P. K.Grover, G. D. Shah, and R. D. Shah, J. Cliern. SOC.,1955,3982;J . Sci. Itid. Res., 15B, 629 (1956). B. R. Samant and A. B.Kulkarni, Iridian J. Clteni., 7,463 (1969). P. R. Iyer and G. D. Shah, Indian J. Chetn.,8, 691 (1970). Y.S.Agasirnundin and S. Rajagopal, J. Org. C/ieni., 36,845 (1971). B. Jackson, H.D. Locksley, and F. Scheinmann, J. Clieni. SOC.(C), 1967,785. B. M.Desai, P. R. Desai, and R. D. Desai, J. Iridiari Clieni. SOC., 37, 53 (1960). M. L. Wolfrom, F. Koniitsky, and J. H. Looker, J . OUR. Client., 30, 144 (1965). H.D. Locksley, A. J. Quillinan, and F. Scheinrnann, Clieni. Conmi., 1969, 1505; J . Clieni. SOC.(C), 1971, in the press. H. D. Locksley, I. Moore, and F. Scheinrnann; J . Clieni. SOC.(C),1966,430. J. S. H. Davies. F. Scheinmann, and H. Suschitzky, J . Org. Clierri.,23, 307 (1958). J. Moron, J. Polansky, and H. Pourrat, Bid/ Soc. chini. Frarire, 1967, 130. E.D. Burling, A. Jefferson, and F. Scheinrnann, Tetraliedoti,21, 2653 (1965). D.H.R. Barton and A. 1. Scott, J. Clierrr. Sac., 1958, 1767. F. M. Dean, J. Goodchild, L. E. Houghton, J. A. Martin, R. B. Morton, B. Parton, A. W.Price, and N. Somvichien, Terraliedrotr Lerf., 1966,4153. 0.R. Gottlieb, M. Taveira Magalhles, M. Caniey, A. A. Lins Mesquita, and D. de Barros Corr&a, Tetrahedron, 22, 1777 (1966). W.J. Horton and J. T. Spence, J. h i . Clierti. SOC.,80,2453 (1958). G.H.Stout and V. F. Stout, Tefraliedron, 14,296 (1961). R. A. Finnegan and P. L. Bachmann, J. Pliarrrr. Sci., 54,633 (1965). A. A. Goldberg and A. H. Wragg, J. Clieni. SOC.,1958,4227. F. Dallacker and Z. Darn6, Clieni. Ber.., 102,2414 (1969). G.H.Stout and W. J. Baikenhol, Tetraliedrori, 25, 1947 (1969). G . H.Stout, E. N. Christensen, W. J. Balkenhol, and K. L. Stevens, Tetrahedron, 25, 1961 (1969). R.Royer, J.-P. Lechartier, and P. Denierseman, Bull. SOC. rltini. France, 1971,1707. J. R. Lewis, Proc. Clieni. Soc., 1963, 373. J. R. Lewis and B. H. Warrington, J . Cliern. SOC.,1964, 5074. J. E.Atkinson, P. Gupta, and J. R. Lewis, Clieni. Contni., 1968,1386;Tetrohedrori, 25, 1507 (1969). J. E. Atkinson and J. R. Lewis, Cliern. Cotnrft., 1967, 803;J . Cherrt. SOC.(C), 1969,281. J. W.A. Findlay, P. Gupta, and J. R. Lewis, Clierri. Corwi., 1969, 206. R. C. Ellis, W.B. Whalley, and K. Ball, Clretri.Comn., 1967,803. A. Jefferson and F. Scheinmann, Narcwe, 207, 1193 (1965);J. Chettt. SOC.(C), 1966, 175. H.D. Locksley, I. Moore, and F. Scheinrnann, Tetroliedvori,23,2229 (1967).
562 200. 201. 202. 203. 204. 205. 206. 207. 208. 209. 210. 211. 212. 213. 214. 215. 216. 217. 218. 219.
220. 221. 222. 223.
The Synthesis of Oxygen Ring Compounds H. D. Locksley and I. G. Murray, J . C h n . Soc. ( C ) , 1970, 392. U. R. Usgaonkar and G . V. Jadhav, J. Indian Cheni. Soc., 40, 27 (1963). Y. Tanase, J . Pharrn. Soc. Japan, 61, 341 (1941). L. A. Paquette, Tetrahedron Len., 1965, 1291, 3103. M. Guyot and C. Mentzer, Bid. SOC.chitti. Frarice, 1965,2558. J. Santesson and G. Sundholm, Arkiu. Kerni, 30, 421 (1969); J. Santesson and C. A. Wachtmeister, Arkiu. Kerrii, 30, 449 (1969). K. R. Markham, Tetrahedron, 21, 1449 (1965). A. C. Jain, V. K. Khanna, and T. R. Seshadri, Citrretrr Sci., 37,493 (1968). Reference 2, Chapters 6-12. Y. S. Agasimundin and S. Rajagopal. Clrern. Ber., 100, 383 (1967). F. Lamb and H.Suschitsky, Tetrahedron,5,1 (1959); J. S. H. Davies,J. Clietn. SOC., 1956,2140; 1958, 1790. P. E. Knott and J. C. Roberts, Plryrocheni., 6, 1597 (1967). V. K. Bhatia and T. R. Seshadri, Tetralredron Lett., 1968, 1741. A. C. Jain, V. K.Khanna, and T. R. Seshadri, Tetrahedron, 25,2787 (1969). M. I,. Wolfrom, E.W. Koos, and H. B. Bhat, J. Org. Cheni., 32, 1058 (1967). A. Jefferson and F. Scheinmann, J. Chern. Sac. (C), 1966, 175. Reference 2, Chapters 6, 8, 9, 10, and 12. B. Jackson, H. D. Locksley, I Moore, and F. Scheinmann, J . Chern. SOC.(C), 1968,2579. J. R. Lewis and J. B. Reary, J . Cheni. Soc. (C), 1970, 1622. H. 1).Locksley, I. Moore, and F. Scheinmann, J . Chem. Soc. (C),1966,2265. A. J. Quillinan and F. Scheinmann, Chem. Cotrirn., 1971, 966. A. A. Lins Mesquita, D. de Barros C o d a , 0. R. Gottlieb, and M. Taveira MagalhBes, Air. Acad. b r a d . Cleric., 1968,40; Anal. Clritn. Acta, 42, 311 (1968). H. D. Locksley, I. Moore, and F. Scheinmann, J. Chetn. Soc. (C), 1966,2186. D. Barraclough, H. D. Locksley, F. Scheinmann. M. Taveira MagalhBes, and 0. R. Gottlieb. J. Cheni. Soc. ( B ) , 1970, 603.
Total Synthesis OfNatural Products, Volume1 Edited by John Apsimon Copyright © 1973 by John Wiley & Sons, Inc.
Subject Index Page numbers in bold face indicate the synthesis of this type of compound. Acetal carbonium ions, 517,518 Acetal hemiacetals, 491,493 cleavage, 491 racemisation, 491 Acetals, 487 amination, 53 dealcoholysis, 41 halogenation, 43,53 synthesis from a-bromo ethers, 376,491 a-chloro ethers, 369,372,377 Acetate pyrolysis, 71 to give a,&unsaturated esters, 345 vinyl ethers, 493,495 Acetate transfer, of hydroxy-esters, 357 Acetic acid, as condensing agent, 524 Acetic anhydride, as condensing agent, 501, 505,509,526 Acetonides, cleavage in the presence of mesylates, 398 as protecting group for 1,2-diols, 13, 19, 72,73,369,377,398 1,3 diols, 360-363, 369, 390 ketones, 94-1 38 selective hydrolysis, 138,400 Acetoxylation, with lead acetate, 527-528, 530,532 of phenols, 527428,530,532 2-Acetoxymethyl pyrroles, 160 self-condensation, 161 meso-Acetoxy porphyrins, 177, 187
acetylation, 178 Friedel-Crafts reaction, 178 hydrogenolysis, 177, 187 Acetylation, of meso-acetoxy porphyrins, 178 of amino alcohols, 369, 381 of chlorins, 206 of porphyrins, 173 Acetylenes, acid-catalysed cyclisation, 249 addition with acid chlorides, 125 as carbohydrate precursors, 8-19 condensation with aldehydes, 17 Diels-Alder reaction, 72 as Grignard reagents, 8-19,62 hydrogenation, 8-19,48,62 cis-hydrogenation, 96, 106, 107, 122, 125 trans-hydrogenation, 135 reduction, 13, 14 in synthesis of chromenes, 469,472 Acetylenic carbanions, addition to ketones, 446 alanes, 133, 138 epoxide cleavage, 48, 133, 138 halogen displacement, 12 Acetylenic ethers, 471472 reduction, 471-472 in synthesis of chromenes, 47 1-472 Acetylenic Grignards, with 0-keto enol ethers, 129 with a@-unsaturatedketones, 125 563
564
Subject Index
Acid chlorides, addition with acetylenes, I25 condensation with ethyl malonamate. 359 malonic ester, 268 diazoketone formation, 20 methyl ketone formation, 20 Acrolein, reductive dimerisation, 9 Acronychia baueri, 484 Acrylic pyrroles, hydrogenation, 155 Actinomycin D.404 Activated esters, in synthesis of peptides, 409-426 Active methylene compounds, see Carbanionic reactants Acylation, of polyalcohols, 398,453 N-Acylation, of methylene pyrrolidones, 235 Acyl azides, reduction, 458 2-Acyl cyclohexanones, stereospecific hydrogenation, 453 Acylium ions, in synthesis of aryl ketones, 429,442 xanthones, 535-539, 546 P-Acyl lactones, rearrangement, 490-491 N 0 Acyl migration, 426 4-Acyloxy-5,6-dihydro4H-pyran 6-carboxylic acid. 4 0 glycal-pseudoglycal rearrangement, 40 2-Acyloxy-5,6-dihydro-2H-pyran6-carboxylic acid, 4 0 glycal-pseudoglycal rearrangement, 40 2-Ac~loxymethylpyrroles, coupling with Grignard reagents, 162 pyrroles, 162 Acyl vinyl amines, photochemical rearrangement, 236 Adam’s catalyst, 25 1,4-Additions, of amides to a,P-unsaturated aldehydes, 345; see also Diels-Alder reaction Adenosyl triphosphate, as phosphorylating agent, 7 Aflatoxins, 485-498 Ageratum, conyyoides, 469 Aglycon, 29 Akabori reaction, 59,60 Alanes, alkynylation, 133, 138 Alcohols, alkyl halide formation in base, 431 amine formation, 343, 376, 378,400,402 -+
azide formation, 343 inversion of configuration, 4, 118, 360363, 378,391,400,447 nitroalkane formation, 382 oxidation, 384, 385 protection as, and regeneration from 0benzoyl propionyl esters, 289 benzyl ethers, 24,50, 137, 369, 372, 422,439,447,448 I-butyl ethers, 128 carbonate esters, 398,448 dihydrocinnamoyl esters, 290 ethers, 432,439 pivaloyl esters, 288 silyl ethers, 134, 136 tetrahydropyranyl ethers, 96-1 38, 291 trityl ethers, 400, 402 reduction, 378 selective benzoylation, 384,402 esterification, 398,453 mesylation, 391 oxidation, 4. 368, 369 protection, 400,432,447 tosylation, 376, 378, 382, 395 tritylation, 400, 402 synthesis by reduction of acyl azides, 458 esters, 458 hydrazides, 458 Alcoholysis, of 2-alkoxy-5.6-dihydro-2Hpyran 6-carboxylic acid, 40 of diazoketones, 20 of dihydro4H-pyrans, 429 of vinyl ethers, 429 Aldehydes, acylation, 524 Akabori reaction, 59 alkyl halide formation, 257 Cannizzaro reaction, 6 condensation, 2-8, 14, 17 with o-amino phenols, 439 carbanionic reactants, 345 phenols, 487 Gatterman synthesis, 527,532 Grignard reactions, 8-19 hydroboration, 43 oxidation by bromine, 4 , 8 polymerisation with pyrrole, 150 protection as and regeneration from thiazolones, 362 Rcformatsky reaction, 18 separation of a-epimers, 362
Subject Index Strecker reaction, 338 synthesis by hydrogenation of nitriles, 60, 232 partial reduction of esters, 439 in synthesis of thiazolidines, 341 Aldimines, 527 condensation, 527 in synthesis of aldehydes, 527 Aldolase, in enzymic synthesis of carbohydrates, 64 Aldol condensation, acid-catalysed, 99 basecatalysed, 1-138 enzyme-catalysed, 64-66 stereospecific, 93, 101 in synthesis of chromenes, 471-480 cyclopentanols, 93,97, 100, 101 Alkaline phosphatase, in enzymic synthesis of oligonucleotides, 319 2-Alkoxy-3-alkylthiotetrahydrofurans, 34,
565
with ethylN-t-butyl malonamate, 351 with ethyl orthoformate, 529-534 with a-halo ethers, 548 of p-keto esters, 132 of ketones, 102, 104, 105, 107, 115,453 of methine carbons, 270 of polyphenols, 487,529434,535, 546, 547 of quinones, 487 of thioamides, 363 of 0,yunsaturated acids, 267 of aJ-unsaturated aldehydes, 113 see olso Carbanionic reactants C-Alkylation, of phenols, 473475 3-Alkyl-5-carboxyalkyl-2,4-dimethyl pyrroles, 152 4-Alkyl-2-carboxyalkyl-3-methyl pyrrole, 153 Alkyl halides, azide formation, 53 35 condensation with indole-magnesium salts, 249 2-Alkoxy-3-alkylthiotetrahydropyrans, 34, hydrogenolysis, 70,368, 378 35 2-Alkoxy-5,6-dihydro-6-hydroxymethyl-2H- reduction, 108, 112 synthesis from alcohols in base, 431 pyrans, 39 aldehydes, 257 2-Alkoxy3,6-dihydro-2H-pyran 6-carboxmeso-Alkyl pyrromethanes, 197 ylic acid, 39 Alkyl sulphenyl chlorides, addition to olealcoholysis, 40 haloalkoxylation, 40 fins, 38 to vinyl ethers, 34, 35 2-Alkoxy-5,B-dihydro-ZH-pyran6,6-dicarboxylic acid, 39 Allenes, as intermediates in synthesis of alcoholysis, 40 chromenes, 471 Diels-Alder reaction, 471 haloalkoxylation, 4 0 Ally1 ethers, Claisen rearrangement, 4712-Alkoxy-5,6-dihydro-2H-pyrans, 35 as precursors of carbohydrates, 35 472, 508-509,513,551-554 rearrangement on silica gel, 553-554 2-Alkoxy-3-hydroxytetrahydropyrans,36 in synthesis of coumarlns, 471472 Alkoxymethylenes, as protecting group for Allylic alcohols, epoxidation, 4 3 cis-diols, 291 oxidation, 125, 231,446 2-Alkoxymethyl pyrroles, coupling with selective oxidation, 118 Grignard reagents, 162 synthesis from epoxides, 43 pyrroles, 162 Alkyl ally1 quinones, oxidative cyclisation, by reductive cleavage of a-keto epoxides, 476 23 1 Alkylation, of amino alcohols, 359, 363 by reductive rearrangement of a,@-unsatof aryl bromides, 265 urated ketones, 231 with benzyloxymethylchloride,270 Allylic thiols, cyclisation with vinyl amines, of carboxylic acids, 351, 355 347 with dimethylsulfoxonium methylide, in synthesis of thiazines, 347 AUyl vinyl ether, Claisen rearrangement, 18 509-5 12 Alumina, as dehydrating agent, 29 with dimethyl sulphate, 487 via enamines, 453.505-506 Aluminium chloride, as condensing agent,
566
Subject Index
488,491,537-539 Aluminium chloride/MeCN, in the selective cleavage of anisoles, 509 Amides, 404426,438 l,4-addition with ad-unsaturated aldehydes, 345 cleavage in the presence of esters, 270, 409426 condensation with a-cyano esters, 235 conversion to imino ethers, 234-261 vinyl amines, 234-261 hydrogenolysis, 369 mild hydrolysis, 369, 372, 378, 379, 393 as protecting group for amines, 369, 378. 385 protection as and regeneration from N-rbutylamides, 355 self-condensation, 234-245 synthesis from lactones, 246 by mixed anhydride method, 381 from polyamines, 403 from thiolactones, 258 in synthesis of vinylogous amidines, 23426 1 corrins, 234-261 see olso Lactams; Piperidones; Pyridones; Pyrrolidones; Pyrrolinones; and Pyrromethane amides Amido keto esters, as carbanionic reactants, 363 Amination, of acetals, 53 of vinyl ethers, 52 Amines, 343, 368, 369, 382, 384,456 addition to methyl isocyanate, 380 a,@-unsaturatedacids, 351 N-bromination, 98 conversion to ketones, 98 deactivation by phthalimide formation. 359 mild hydrolysis, 230 protection as and regeneration from acetamides, 369, 378 benzamides, 355 carbamates, 343, 369, 372, 381, 394, 409426 dimethylaminoimines, 288 dinitrophenyl amines, 369. 372, 376 isobutyl carbamic esters, 288 phthalimides, 409426 thioamides, 363
tosyl amides, 409426 trityl amines, 341,347,409426 reductive methylation, 381 selective acylation, 403 selective methylation, 355,359, 363 synthesis from alcohols, 343, 376, 378, 400,402 by hydrogenation of aides, 341,403 oximes, 377 by hydrogenolysis of N-alkyl carbamic esters, 225 urea formation, 380 see also Amino sugars I-Amines, condensation with a-diketones, 154 coupling with malonic ester, 158 as intermediates in annelation, 158 a-Amino acids, 338 deamination, 24 protection as and regeneration from oxazolones, 338 P-Amino acids, p-lactam formation, 337-348 Amino alcohols, 391 acetylation, 369, 381 alkylation, 359, 363 Aminoalkoxylation, of vinyl ethers, 52 Amino benzoquinones, 455457 2-Amino crotonate, condensation with ahaloacetaldehyde, 15 1 a-haloketones, 15 1 2-Amino-3-cyano pyrroles, cyclisation with formamidine acetate, 400 Aminocyclopentanones, 249 Aminodeoxyinositols, 66-75 a-Amino esters, 35 1 as carbanionic reactants, 339, 341 condensation with ad-unsaturated esters, 2 30 vinyl ethers, 230 a-Amino ethers, 393,394,398,400 Amino group, as leaving group, 158. 199 Aminohydroxylation, of olefins, 47,48,49, 60 a-Aminoketones, condensation and cyclisation with carboxylic acid derivatives, 227 with 0-keto esters, 150 synthesis from oximinones, 150 in synthesis of pyrrolinones, 227 Wittig reaction, 439
Subject Index &Amino levulinic acid, enzymic incorporation into porphobilinogen, 156 Aminomethyl pyrroles, 157 o-Amino phenols, condensation with alde hydes, 439 oxidative coupling, 422 protection as and regeneration from oxazolines, 439440 Amino sugars, 58-60,364-404 acetylation, 369 separation as imines, 391 a-Amino thiols, cyclisation to thiazolidines, 339,341 protection as and regeneration from thiazolidines, 342-345 p-Amino tosylates, 385 aziridine formation, 385 Ammonium mucate, pyrolysis, 150 Ammonolysis, of esters, 368 Amorphin, 498 Amphipyrones, 525-534 cleavage, 525-534 isoquinoline formation, 527 Amphotericin B, 427 Anhydronodakenetin, 478 Anilidates, as protecting group for phosphate, 293 Anilinium chloride, as demethylating agent, 515 Anisolcs, clcavagc with boron tribromidc, 509,546 boron trichloride, 551 piperidine, 551 reduction to 1,4-dienes, 251 selective cleavage, 509, 515, 521, 524, 5 37-55 2 Anomeric centre, 2, 29 control of epoxide cleavage, 36 polar effects, 36 Anomeric effect, 30, 38,47 Anomerisation, 2 Anthraquinones, 496498 Antibiotics, 33 1-458 sugar moiety, 47 Arachidonic acid, as precursor of prostaglandins, 84 Arndt-Eistert procedure, in homologation of carboxylic acids, 246 Aromatic compounds, as carbanionic reactants, 351, 355
567
vinylation, 476 Aryl aldehydes, 507-509; see olso Benzaaldehydes Aryl amines, 438,448 Aryl esters, Dieckmann condensation, 444 Fries rearrangement, 442 see also Coumarins Aryl ethers, 467-555 cleavage with boron tribromide, 509, 530, 546 boron trichloride, 551 carbanionic reactants, 496 neighbouring group participation, 515, 521,524,537-543 reduction, 115,251,265.441 selective cleavage, 487,509,515, 521, 524, 525.537-552 see also Anisoles; Benzyl ethers; Chromans; Chromenes; Coumaranones; Ethers; and Xanthones Aryl halides, displacement of chlorine, 355, 444 hydrogenolysis, 351, 355,444 Arylidene pyrrolinones, 219 Aryl ketones, 351,355,429,441,442,467555 coupling with phenols, 5 18 hydrogenolysis, 35 1 reduction, 476-479, 533 see also Chromanones; Chromenones; Coumaranones; and Xanthones Ascochyta fabae Speg., 528 Ascochyta pisi Lib., 528 Aspergillus flavus, 486 Aspergillus parasiticus, 486 Aspergillus versicolor (Vuillemir) Tiraboschi, 485,496 Aversin, 486 Azaindoles, 158 Mannich reaction, 158 p-substitution, 158 Azaphilones, 525 Azides, reduction, 53,235, 341, 343, 376, 378,403,456 synthesis from alkyl halides, 53 in synthesis of peptides, 409-426 Azidoketene, 34 I addition with imino thio ethers, 341 ‘thiazoles, 341 thiazolines, 341
568
Subject Index
in synthesis of&lactams, 341 Aziridines, 235,385,455457 cleavage, 235, 385,456 Azlactones, 355,499 in synthesis of benzyl cyanides, 499 tetracyclines, 355 Azocarboxy ester, 343 as leaving group, 343 Azo compounds, 448 hydrogenation, 448 Bacillus brevis, 406 Bacillus polymyx, 4 13 Bacillus prodigiosus, pyrrole content, 227 Bacitracins, 404 Bacteriochlorins, 209 Baeyer-Villiger reaction, 73 of cycloalkyl ketones, 116 of cyclobutanones, 92, 112, 113 of norbornenones, 108 Baryta, as condensing agent, 2 Beauvericin, 406 Beckmann rearrangement, of oximes, 254 in synthesis of piperidones, 254 Benzaldehydes, 507-509 Dakin reaction, 547 oxidative cleavage, 547 Benzamides, as protecting groups for amines, 355 Benzo-dihydrofurans, 467-555 Benzofurano( 3',2':3,4)coumarins, see Coumestans Benzofurans, 467-555 hydrogenation, 522 Benzoic acids, 528,532 Benzoic esters, hydrolysis in the presence of acetonides, 384 Benzophenones, 441,442 Benzopyrans, 467-555 hydrogenation, 522 Benzoylation, of polyalcohols, 384,402 Benzoylbenzoic acids, as intermediates in synthesis of anthraquinones, 496498 P-Benzoylpropionyl esters, as protecting group for alcohols, 289 Benzyi alcohols, oxidation, 501, 504 Benzylamines, hydrogenolysis, 25,60, 372 Benzylation, of polyphenols, 449, 487 of quinones, 487
Benzyl cyanides, 499 Hoesch condensation, 500,503 Benzyl esters, hydrogenolysis, 117, 155,185 as protecting group for carboxylic acids, 419,429 Benzyl ethers, hydrogenolysis, 24, 50, 135, 138 mild hydrogenolysis, 493 as protecting group for alcohols, 24, 50, 137,369,372,422,439,447,448 hemiacetals, 493 phenols, 487,493,518,530 selective cleavage, 525 see also Aryl ethers; Ethers Benzylidenes, as protecting group for cisdiols, 29 1 1,3-diols, 378 Benzyloxymethyl chloride, as alkylating agent, 270 BicycloI 2.2.1 1heptene, as protecting group for trans-olefins, 130 retro-Diels-Alder reaction, 131 Bicyclo[ 3.1.01 hexanes, conformational analysis, 92, 93 stereochemistry of cleavage, 99, 103-107, I12 a,c-Biladienes, 179-184 cyclisation, 179-184 in synthesis of porphyrins, 179-184 a-Bilane-1'8'-dicarboxylic acids, cyclisation, 185 oxidation, 185 Bilanes, acid catalysed rearrangements, 185 b-Bilene-l'8'-dicarboxylicacid, condensation and cyclisation with trimethyl orthoformate, 185 b-Bilene carboxylic acids, decarboxylation, 195,197 Bilenes, 209-227 b-Bilenes, 194-199 condensation and cyclisation with trimethyl orthoformate, 195, 197 hydrogenation, 195 reactivity of 6-carbon, 197 in synthesis of porphyrins, 194-199 Bile pigments, 209-227 Bilirubic acid, oxidation, 213 Bilirubin, mild reduction, 21 3 2,2'-Bipyrroles, 227-232 Birch reduction, of aryl ethers, 115, 251
Subject Index of chromenes, 472 Boron tribromide, in cleavage of aryl ethers,
509,530,546
Boron trichloride, in cleavage of aryl ethers, 55 1 Bourgel's catalyst, 10 Branched-chain sugars, 62-64 Bromine, as oxidising agent, 4,8 as dehydrogenating agent, 533 Bromine/sulphuryl chloride, as oxidising agent, 165 Bromobenzenes, alkylation, 265 4-Bromo-2-methylbut-2-ene,in alkylation of phenols, 473474 2-Bromomethylpyrroles, self-condensation,
161
N-Bromosuccinimide, as brominating agent,
18,72,345,550
in bromination of amino group, 98 as oxidising agent, 441 N-t-Butyl amides, as protecting group for amides, 355 t-Butyl esters, mild hydrolysis, 189,195 t-Butyl ethers, as protecting group for alcohols, 128 t-Butyl hypochlorite, as halogenating agent,
185
as oxidising agent, 185 Cadmium dimethyl, in synthesis of methyl ketones, from anhydrides, 263 from acid chlorides, 263 Calcium carbonate, as condensing agent, 2 Calcium hydroxide, condensing agent, 2,s Calophyllum brasiliense Camb., 549 Calophyllum sclerophyllum Vesq., 549 Cannabis sativa, 481 Cannizzaro reaction, 6 Capreomycin, 404 Carbamates, hydrogenolysis, 225 as protecting group for amines, 343,369,
372, 381, 394,409-426
Carbanionic reactants, acetylenic carbanions,
12,48,133,138,446
alanes, 133,138 aldol condensations, 1-138,471-481 amido keto esters, 35 1, 363 a-amino esters, 339,341 aromatic compounds, 351,355 1,3-bis(thiomethyl)allyl lithium, 1 1 3
569
carbomethoxymethyl chlorins, 206 a-cyano esters, 235 cyanomethyl benzenes, 496 cyclopentadienyl anion, 108 dithioacetals, 97 enolates, 130 in epoxide cleavage, 113, 133, 138 esters, 349-351 ethyl-N-r-butyl malonate, 35 I ethyl malonamate. 359 Grignard reactions, 8-19,62,125, 129,
162,165,199,356,360,469,527
0-keto enol ethers, 120 P-keto esters, 132, 150, 151, 155,351 ketones, 102-107,115,250,351 lithium aryl, 478-484 malondialdehyde, 345 malonic ester syntheses, 16, 158,230,268 4-methyl pyridines, 158 nitroalkanes, 100. 101,238,369,382 a-nitro esters, 357 n-pentyl lithium, 113 pyrroles, 143-271 pyrrolinones, 219 reaction with a#-unsaturated aldehydes,
113
thio ethers, 343 Carbenoid reagents, addition to olefins, 102,
112
cyclopropane formation, 102,112 1,l-dichloroacetone, 112 ethyl azidoformate, 102 Carbohydrates, 1-75.364-386 t-Carbomethoxy groups, extrusion, 351 Carbomethoxylation, of pyrroloindoles, 456 Carbonate esters, as protecting group for alcohols, 398,448
1,2-diols,67,68,448
phenols, 441 stability to acid, 398 Carbonyl compounds, selective Grignard reaction, 356, 360 selective reduction, 35 1 N,N'-Carbonyldiimidaole,as condensing agent, 422 5Carboxyalkylpyrrole-2-carboxylicacids,
159
decarboxylation, 159 Carboxylation, Kolbe reaction, 532 of phenols, 528,532
570
Subject Index
with potassium bicarbonate in glycerol, 528 Carboxyl group, as leaving group, 172, 177, 185, 195,214,263 stabilising effect in bilanes, 185 Carboxylic acids, alkylation, 355 via mixed isopropyl carbonic anhydrides, 351 Arndt-Eistert reaction, 246 condensation and cyclisation with aamino ketones, 227 homologation, 246, 355 oxidative decarboxylation, 117 protection and regeneration, 409-426 from p-methoxybenzyl esters, 345,429 from p-nitrobenzyl esters, 345 from trichloroethyl esters, 89, 345 reduction, 4, 18 in synthesis of pyrrolinones, 227 5-Carboxyl-5'-methylpyrromethenes,condensation with S-brorno-S'-bromomethyl pyrromethenes, 172 Cardiac glycosides, sugar moiety, 49 Catechols, 1.3-dioxole formation, 518, 522 oxidative coupling, 516-521 protection as and regeneration from 2,2diphenyl-1,3-dioxoles, 519 Wanslick coupling, 516-521 see olso Phenols Cellulose TLC, of oligonucleotides, 298 Chalcomycin, 426 sugar moiety, 53 Chalcones, oxidation, 51 2 Chlorins, 199-209 acetylation, 206 formylation, 206 reduction, 209 Chlorocruoroheme, porphyrin moiety, 189 Chlorohydrins, from vinyl ethers, 32 N-Chloromercury compounds, coupling with a-chloro ethers, 393, 394, 398 1Chloro-1-nitrosocyclohexane,Diels-Alder reaction, 25 in synthesis of amino sugars, 25 in synthesis of 1,2-oxazines, 25 mChloroperoxy benzoic acid, as oxidising agent, 345,433 Chromanochromanones, dehydrogenation, 499 Chromanochromans, 498-5 13
Chromanols, oxidation, 501, 504 Chromanones, 476,505-507,533 condensation with phenols, 506-507 quinones, 5 12 cyclisation to chromenes, 476 enol acetate formation, 505 reduction, 476,533 Chromans, 473481 alkylation, 505-506 Chromatography, of oligonucleotides, see Oligonucleotides, chromatography Chromenes, 468485,552 acid-catalysed rearrangement, 469,478 base-catalysed rearrangement, 483 oxidative ring contraction, 517, 518 reduction, 472 see dso Pyranoxanthones Chromenochromones, 498-5 13 cleavage, 499 hydrogenation, 501 reduction, 501-502 Chromenones, see lsoflavones Chymotrypsin, in specific cleavage of dihydrocinnamoyl esters, 289 Cinerubin A, sugar moiety, 56 Claisen condensation, of deoxybenzoins, 514,515 with ethyl orthoformate/piperidine, 512 ethyl oxalate, 5 12 of a-fluoro esters, 61 in synthesis of isoflavones, 512 Claisen rearrangement, of ally1 ethers, 471472, 508-509, 513, 551-554 of ally1 vinyl ether, 18 stereospecificity, 508-509 Cobyric acid, 232-268 Condensing agents, acetic acid, 524 acetic anhydride, 501.505, 509, 526 aluminum chloride, 488.491, 537-539 baryta, 2 calcium hydroxide, 2, 5 calcium carbonate, 2 N,"-carbonyldiimidazole, 422 cuprous chloridelpyridine, 495 I-cyclohexyl-3[ 2-morpholinyl-(4)-ethyl] carbodiimide, 412 dicyclohexyl carbodiimide. 341, 345, 348, 404,409426,504-506 diethylene glycol, 522 dimethyl sulphoxide, 512
Subject Index
571
enzymes, 64 Curtius rearrangement, 16 ~-ethyl-5-phenylisoxazolium-3'-sulphonate,Cyanoalkanes, oxidation to ketones, 498 422 Cyano anthracenes, 496 magnesia, 2 as intermediates in synthesis of anthramagnesium hydroxide, 3 quinones, 496 mineral clay, 2 a-Cyano esters, 448 oxallyl chloride, 495 condensation with amides, 235 oxallyl chloride/aluminium chloride, 488, Cope condensation, 452 PCyanoketones, from aJ-unsaturated 49 1 phosphorous pentoxide. 530 ketones, 232,262 piperidine, 495, 505 Cyanomethyl benzenes, as carbanionic repolyphosphoric acid, 500,510, 526, 535actants, 496; see also Benzyl cyan. 539 ides potassium carbonate, 14 Cyanomethyl esters, in synthesis of peptides, pyridine, 552 409-412 pyridinium chloride, 542 aCyano tosylates, 385 ?-radiation, 2 aziridine formation, 385 silver oxide, 549 Cyclic amides. see Lactams; Pyrrolidones; sodium hydroxide, 5 and Pyrrolinones tetramethylammonium hydroxide, 541 Cyclitofs, 66-75 thallium hydroxide, 5 1,2-Cycloadditions, of a-halovinyl ketones, trifluoroacetic anhydride, 532 125 Woodward's reagent, 422 photochemical, 125 zinc carbonate, 494 of aJ-unsaturated ketones, 125 zinc chloride/phosphoryl chloride, 535see olso Diels-Alder reaction 1,4-CycIoadditions, see Diels-Alder reaction 9,548 Cope condensation, of a-cyano esters, 452 Cyclobutanes, 483 ofp-keto esters, 452 Cyclobutanones, 91, 112, 113 Cope reaction, 60 Cyclobutyl ketones, reduction, 125 Copper-nickel alloy, as dehydrating agent, Cyclohexanehexols, 66-75 Cyclohexanepentols, 66-75 29 Copperquinoline, as decarboxylation cataCyclohexenes, ring contraction, 360 Cycloheximides, 452 lyst, 131 Copper sulphate, as dehydrating agent, 476 1-Cyclohexyl-3[ 2-morpholinyl-(4)-ethyl] Corrins, 232-268 carbodiimide as condensing agent, Coumaranones, 444 412 bromination, 491 Cyclopentanols, 81-138 Coumarins, 448,476,488,490495,506, Cyclopentenone prostanoic acids, 125-132 507,s 12-525 Cyclopen tenones, 8 1- 138,488-494 coupling with diazonium salts, 448 Cyclopropanes, 102,509 Crignard reaction, 469 base-catalysed cleavage, 93 partial hydrogenation, 488 stereochemistry of cleavage, 93, 99, 103reduction, 490, 522 107,112 see also Isoflavans Coumermycin A, 435 Dakin reaction, of benzaldehydes, 547 Coumestans, 5 12-525 Daunomycin, 435 Coupled oxidation, of porphyrins, 210 DEAF.-Cellulose, in chromatography of Crombie synthesis, of chromenes, 551 oligonucleotides, 294 Cuprous chloridelpyridine, as condensing Dealcoholysis, of acetals, 4 1 Deamination, of amino acids, 24 agent, 495
-
572
Subject Index
with neighbouring group participation, 24 Decarboxylation, acid-catalysed, 35 1 anodic, 263 of b-bilene carboxylic acids, 195,197 of I-carbomethoxy groups, 351 with copperquinoline, 131 of glyoxylic acids, 496 of a-keto acids, 496 ofp-keto acids, 131 of p-keto esters, 351 of malonic esters, 268 of pyrrole-carboxylic acids, 157, 159, 217, 220, 222,226,230 of a,&?-unsaturated esters, 235, 239 Decursin, acid catalysed rearrangement, 478 Dehydrating agents, alumina, 29 anhydrous copper sulphate, 476 copper-nickel alloy, 29 N,N-dicyclohexylcarbodiimide,34 1, 345, 348,404,409-426 N,N-diisopropylcarbodiimide,341 hydrogen fluoride, 351 lead acetate, 363 phosphorous tribromide/pyridine, 503 polyphosphoric acid, 355 sodium bisulphate, 476 tosyl chloride/collidine, 478 triethyl ammonium acetate, 360 see also Condensing agents Dehydrogenation, with bromine, 533 of chromanochromanones, 499 with dichlorodicyanobenzoquinone,472475 of dihydroamphipyrones, 533 of o-dimethylally1 phenols, 471-475 of 1,4-diols, 29 with cobalt catalyst. 29 with iodinelsodium acetate, 499 with manganese dioxide, 533 with manganese dioxide/ferricyanide, 499 with mercuric oxide, 533 of pyrrolidinylpyrroles. 230 of pyrrolylmethylpyrrolinones, 217 with selenium dioxide, 441 of a,&?-unsaturated ketones, 441 see also Oxidation Dehydrohalogenation, basecatalysed, 40, 42,43, 71 of bromo ketals, 231 with 1,5diazabicycb(4,3,0)-non-S-
ene, 23 I of vic-dihalides, 32, 34 Dehydropterocarpans, 522 Demethylation, with aluminium chloride/methyl cyanide, 509 with anilinium chloride, 516 with boron tribromide, 509 with hydrogen bromide, 514 of methylpyrromethenes, 21 3 of polymethoxybenzenes, 509, 51 5, 521, 524,537-552 Denitration, of nitrobenzenes, 540 Deoxybenzoins, Claisen condensation, 5 14,
515
Deoxyfluoro sugars, 60-61 Depsipeptide antibiotics, 404-4 26 Derric acid, 503 Derris dust, 498 1 1-Desoxyprostaglaidins, 124-125 2,4-Diacetyloxophlorins, reduction, 178 Dialkoxylation, of 1,3-dienes, 56 1,S-Diazabicyclo-(4.3.0)-non-5-ene, as a dehydrohalogenating agent, 231 Diazoketones. 20,394 acetolysis, 22 from acid chlorides, 20 alcoholysis, 20 diethyldithioacetal formation, 20 hydrolysis, 394 in synthesis of bromomethylketones, 530 Wolff rearrangement, 20 Diazomethane, diazoketone formation, 20 epoxide formation, 22 halogen displacement, 444 reaction with acid chlorides, 20 ketones, 22 Diazonium salts, coupling with coumarins, 448 Dibenzopyrans, oxidation, 521 1,3-Dibrorno-5,5-dimethylhydantoin, in bromomethoxylation of olefins, 38, 41 5,5'-Di(bromomethyl)pyrromethenes,condensation with pyrromethanes, 169 1,4-Dicarbonyl compounds, in synthesis of pyrroles, 232; see also y-Diketones 2,5-Dicarboxyalkylpyrroles,160 Dicarboxylic acids, 265 selective esterification, 124 5,s'-Dicarboxylpyrromethanes,condensa-
Subject Index
573
tion with S,S'diformylpyrroketones, of vinyl ethers, 30,41 refro-Diels-Alder reaction, of norbornenes, 177 formic acid, 167 131 Dienediones, selective reduction, 131 1,l-Dichloroacetone, as carbene precursor, 1,3-Dienes, dialkoxylation, 56 112 2,3-Dichloro-5,6-dicyano-l,4-benzoquinone, Diels-Alder reaction, 39,40, 67-72, 93, as dehydrogenating agent, 472-475 97,108,236,245,356,360 frans-hydroxylation, 6 7 in oxidation of allylic alcohols, 118 1,4-Dienes, synthesis by reduction of aryl Dichlorodicyanoquinone, in oxidative couethers, 25 1 pling, 544.552 Dienones, selective oxidative cleavage, 124 Dichlorodiphenylmethane,in synthesis of 2,2diphenyl-l,3-dioxoles,519 2,2-Diethoxyalkylamines,condensation and Dichloromethyl methyl ether, in formylacyclisation with adiethylphosphontion of chlorins, 206 alkanoic acids, 227 in synthesis of pyrrolinones, 227 N,"-Dicyclohexylcarbodiimide, as conDiethyl dithioacetals, from diazoketones, 20 densing agent, 504-506 hydrogenolysis, 20 as dehydrating agent, 341, 345, 348,404, in synthesis of methyl ketones, 20 409426 gem-Dlcyanovlnylpyrromethanes,mild oxiDiethylene glycol, as condensing agent, 522 dation, 203 Diethyl-1-octynylalane, alkynylation, 133, 2,3-Dideoxy-DL-2-enopyranos-4-uloses; 56 138 Diethyl oxalate, condensation with 4-methyl2,3-Dideoxytetrofuranose,equilibrium with 1,4-hydroxyaldehydes, 28,29 pyridines, 158 Dieckmann condensation, of aryl esters, 444 a-Diethyl phosphonalkanoic acids, condensation and cyclisation with 2,2in synthesis of 0-diketones, 254 pyrroles, 230 diethoxyalkylamines, 227 Diels-Alder reaction, of acetylenes, 72 in synthesis of pyrrolinones, 227 of allenes, 471 S,S'-Diformylpyrroketones, 177 in carbohydrate synthesis, 1-75 condensation with 5,5'dicarboxylpyrroof 1-chlorel-nitrosocyclohexane,25 methanes, 177 pyrromethanes, 177 of I,j-dienes, 39,40, 67-72, 93, 97, 108, S,S'-Diformylpyrromethanes, condensation 236,245,356,360 with pyrroles, 183 of dithio esters, 27,28 pyrromethanes, 168 of furan, 67 vic-Dihalides, dehydrohalogenation, 32,34 in heterocycle synthesis, 24-28 intramolecular, 554 halogen displacement, 38,39 of a-keto esters, 39, 40 5,s'-Di(halomethyl)pyrromethenes, condensation with 5,S'dibromopyrromethof ketones, 39.40 of olefins, 30,236,356, 360,483 enes, 170 stereospecificity, 97 S,S'-Dihalopyrrome thenes, condensation in synthesis of oxazines, 25 with 5,5'-di(bromomethyl)pyrrw of up-unsaturated aldehydes, 30,41,43 methenes, 170 of apunsaturated esters, 30, 70-72, 236 5,5'-dlmethylpyrromethenes. 170 of a$-unsaturated?-keto esters, 245 Dih ydroamphip yrones, dehydrogenation, 533 of ad-unsaturated ketones, 483 Dihydrocinnamoyl esters, as protecting of ap,y,b-unsaturated ketones, 554 of ap-unsaturated nitriles, 30, 93, 108 group for alcohols, 290 Dihydrocoumarins, 488 of a,&unsaturated nitro compounds, 97 hydrolysis, 490 of vinylcnc carbonate, 67,68 rearrangement, 490 of vinyl esters, 70, 71
574
Subject Index
2,3-Dihydrofurans, acetal formation, 30, 31 addition with quinones, 487 as carbohydrate precursors, 28-58 Dihydropyranones addition of, NOCI, 377 3,4-Dihydro-2H-pyrans, 468485 acetal formation, 30, 31 as carbohydrate precursors, 28-58 DihydrdH-pyrans, 429 alcoholysis, 429 Dihydrothiains, 27,28 Dihydrothiopyrans, 27,28 2,2'-Dihydroxybenzophenones,535-543 as intermediates in synthesis of xanthones, 535-543 Dihydroxychlorins, 201 Pinacol rearrangement, 201 5,7-Dihydroxymethylcoumarin,in synthesis of aflatoxins, 487490 1,3-Dihydroxyxanthones, hydroxylation, 547-549 substitution, 547-549 A',"-Diisopropylcarbodiimide, as dehydrating agent, 341 Diketene, in synthesis of pketo esters, 531 a-Diketones, condensation with t-amines, 154 0-Diketones, 254 base-catalysed cleavage, 444, 499 condensation with oximinones, 152, 157 piperidones, 254 oxidative cleavage to 1.5-dicarboxylic acids, 265 protection as and regeneration from isooxazoles, 26 1-268 in synthesis of N-enonepiperidones, 254 7-Diketones, 453 cleavage, 36 1 conversion to polyenones, 120 cyclisation to pyrroles, 150 pyrrolinones, 236 stereospecific hydrogenation, 453 1,2-Dimesylates, epoxide formation, 403 o-Dimethylallylphenols, 471475 cyclisation to chromenes, 472475 2-N,N-Dimethylamidopyrroles, complexation with phosphoryl chloride, 165 coupling with pyrroles, 165 Dimethylaminoimines, as protecting group for amines, 288 2-Dimethylaminomethylpyrroles,self-condensation, 199 Dimethylazodicarboxylate,in substitution
of thiazolidines, 343 1',8'-Dimethyk,c-biladienes. oxidative cyclisation, 183
NjV-Dimethylcarboxamidopyrroles,160
2,2-Dimethyl-2H-chromene, 468-485 rrans-2,4-Dimethylcyclohexanones, 453 2,6-Dimethyl-4-pyrone, 527 5,5'-Dimethylpyrroketones,oxidation, 177 5,5'-Dimethylpyrromethenes,condensation with 5,5'-dibromopyrromethenes, 170 Dimethyl sulphate, as alkylating agent, 487 Dimethylsulphoxide. as condensing agent, 512 Dimethylsulphoxonium methylide, as alkylating agent, 509-512 Dinitrophenylamines, as protecting group for amines. 369, 372, 376 1,2-Diols, dehydration to ketones, 114 epoxide formation, 403 8-hydroxy ether formation, 31 olefin formation, 403 oxidative cleavage, 8, 15,73, 97, 116, 117,343,360,448 protection as and regeneration from acetonides, 13, 19, 72, 73, 369, 377, 398 carbonates, 61,68,448 selective alkylation, 138 see also Catechols rrans-l.2-Diols, 209 1,3-Diols, ester migration, 453 protection as and regeneration from acetonides, 360-363, 369, 390 benzylidenes, 378 selective acetylation, 453 stereospecific synthesis, 453 1,4-Diols, dehydrogenation, 29 cis-Diols, protection as and regeneration from alkoxymethylenes, 291 benzylidenes, 291 Dioxolanes, as protecting groups for ketones, 355 1,3-Dioxoles, 518,522 as protecting group for catechols, 519 2,2-Diphenyl-l,3dioxoles, as protecting group for catechols, 519 Dipyrrylketones. see Pyrroketones Dipyrrylmethanes, see Pyrromethanes Dipyrrylmethenes, see Pyrromethenes
Subject Index Dithianes, as protecting group for ketones,
97
Dithioacetals, 385 in ketone synthesis, 97 Dithio esters, Diels-Alder reaction, 27,28 ap,a',p'-Diunsaturated ketones, hydrogenation, 441,442 D.N.A. Polymerase, in enzymic synthesis of oligonucleotides, 314-316 Dothistromin, 486 Duff condensation, in formylation of phenols, 507-509,547 Edward-Lemieux effect, 30
$8.1 1,14.17-Eicosapentaenoic acid, as precursor of prostaglandins, 84
8,11,14-Eicosatrienoic acid, as precursor of prostaglandins, 84 Elliptone, 507-509 Emmons reaction, in synthesis of pyrrolinones, 221 Enamides, addition with HCN, 242 with thiolactams, 241-261 imino-ether formation, 238,239 in synthesis of vinylogous amidines, 238,
239
see also Methylene pyrrolidones
Enamines, 235,239 in acylation of aldehydes, 524 in alkylation of chromans, 505-506 ketones, 453 condensation with imino ethers, 237,239,
243
as intermediates in separation of aepirnerlc aldehydes, 362 stereospecific hydrolysis, 362 in synthesis of xanthones, 546 P-Enamino ketones, 261-268 Enniatins, 405 Enol acetates, of chromanones, 505 directional specificity, 505 Enolates, Wittig reaction, 120 Enol ethers, see Vinyl ethers N-Enone piperidones, 254 Enzymes, as condensing agent, 64 in oxidative coupling, 544 in synthesis of oligonucleotides, alkaline phosphatase, 319 D.N.A. polymerase, 314-316 nucleotidyl transferase, 316
575
phosphodiesterases, 3 19 polynucleotide kinase, 319 polynucleotide ligase, 320-321 polynucleotidyl phosphorylase, 3 17-318 ribonucleases, 318-319 R.N.A. polymerase, 316-317 in synthesis of sugars, 64-66 Episulphides, 241-261 in ring expansion of thiazolidines, 345 cleavage, 345 sulphur extrusion, 241-261 Episulphonium ion, 35, 38 Epoxidation, of allylic alcohols, 43 of a-hydroxytosylates. 138 of ketones, 22 of olefins, 8-19,26,38-49,60 of olefins stereospecifically, 93, 113 stereochemical effects, 36 of ap-unsaturated ketones, 231 of vinyl ethers, 31 see also trans-Hydroxylation, Oxidation Epoxides, alkynylation, 133,138 cleavage by acetylenic carbanions, 48 acid-catalysed hydrolysis, 31, 39 alanes, 133, 138 alcohols, 36,54 amines, 48,49,391 base-catalysed hydrolysis, 39,42,43 carbanionic reactants, 113, 133, 138 formate ion, 27 hydride reagents, 36,45,46 iodine, 402 ion exchange resin, 22 hydrogenation, 398 isomerisations, 43 olefin formation, 403 stereochemistry, 36 synthesis from diazomethane, 22 1,2-dimesylates, 403 halohydrins, 9,10,11 ketones, 22 olefins, 8-19,26,3849,60 vinyl ethers, 3 1 aJ-Epox ycyclopropanes, stereochemistry of cleavage, 103 a$-Epoxyketones, reductive cleavage, 1 1 8 Erythromycin, 426 sugar moiety, 47 Esterification, of polyalcohols, 398,453 Ester migration, in esters of 1,3-diols, 453
576
Subject Index
Esters, ammonolysis, 368 as carbanionic reactants, 349-351 McFadyen-Stevens reduction, 230 mild hydrolysis, 368, 380 mild reduction, 458 reduction with hindered hydride reagents, 439 reduction via hydrazides, 458 selective hydrolysis, 133, 398 selective reduction, 133 stability towards hydride reagents, 230 Ethanolysis, of resorcinols, 531 Ethers, bromination, 49 1 cleavage, 35 1,355, 359 hydrolysis in the presence of lactones, 112 1,3-diols & 0-hydroxyketones, 109 as protecting group for alcohols, 432,439 see also Aryl ethers; Benzyl ethers Ethyl azidoformate, as carbene precursor, 102 EthylN-t-butyl malonarnate, a new alkylating agent, 351 Ethyl malonamate, as carbanionic reactant, 359 N-Ethyl-S-phenylisoxazolium-3'-sulphonate, as condensing agent, 422 Ethyl orthoformate, as an alkylating agent, 528-534 condensation with pyrroles, 219, 222,226 Ethyl orthoformatelpiperidine, Claisen condensation, 5 12 Ethyl oxalate, Claisen condensation, 512 Ethyl thioesters, as protecting group for phosphate, 292 EvodiP littoralis, 475 Ferricyanide, as oxidising agent, 516521,543 Filipin, 427 Flavilium salts, 5 18 oxidation, 517, 518 Flavones, 410 in synthesis of chromenes, 470
Fluoro sugars, 60-61 Formaldehyde, Cannizzaro reaction, 6 condensations, 2-8 with pyrroles, 16 1 formose reaction, 2-8 in interstellar gases, 2 in reductive methylation of amines, 381 Formamidine acetate, in synthesis of pyrrolopyrimidines, 400 Formose, fractionation, 7 Formose reaction, in synthesis of carbohydrates, 2-8 Formylation, of a-amino esters, 339, 341 of chlorins, 206 with dichloromethyl methyl ether, 206 by Duff condensation, 507409,547 Gatterman reaction, 219 with hydrogen cyanide, 219,223 of 0-keto enol ethers, 120 of ketones, 262 of phenols, 507-509, 547 of porphyrins, 182 of pyrroles, 217, 219.223 of pyrrolylmethylenepyrrotinones,21 7 of pyrrolylmethylpyrrolinones,21 7 of resorcinols, 528-534 Vilsmeier-Haack method, 217 of xanthones, 547 meso-Formyl chlorins, 202 4-Formylcoumarins, 488 7-Formyl esters, 6-lactone formation, 202 Formyl group, conversion to methoxycarbonylmethyl, 202 2-Forrnylindoles, pyrroloindole formation, 456 2-Formyl-N-rnethylpyrroles, as coupling agent in pyrromethane synthesis, 168 a-Formylpyrroles, condensation with pyrroles, 143-271 pyrrolinones, 219, 220, 225 Knoevenagel reaction, 155 a-Formylpyrrolylmeth ylenepyrrolinones, 217 condensation with pyrrolylmethylenepyrrolinones, 217 a-FormylpyrrolylmethyIpyrrolinones, condensation with pyrrolylmethylpyrrolinones, 217 a-Formylpyrromethanes, condensation with
Subject Index pyrromethane a-carboxylic acids, 195,197 pyrromethanes, 195,214 pyrromethenes, 214,215 Frasera caroliniensis, 54 1 Friedelcrafts reaction, of meso-acetoxyporphyrins, 178 of porphyrins, 173, 182 of pyrromethenes, 180 in selective substitution, 441 Fries rearrangement, photochemical, 442 with titanium tetrachloride, 442 of aryl esters, 442 Fumagillin, 435 Furan, Diels-Alder reaction, 67 Furanosides, 3 1 Futanoxanthones, 548,553 Furans, 478,507-5 10 as carbohydrate precursors, 28-58 conformational analysis, 28-58 conversion to pyrroles, 154 Furobenzofurans, 486498 Furopyrones, 55 1 cis-Fused rings, 502,504,509 Fusidic acid, 435 Gambogic acid, 468 Garcinia eugenifolia, 545 Gatterrnan reaction, formylation of pyrroles, 219 synthesis of aldehydes, 527, 532 Genes, 321-324 Gentiana belladifolia, 541 Gentianaceae, 534,541 Gentiana lutea, 543 Gibberella zeae, 429 LGlutamic acid, conversion to D-Ribose, 24 Glycal-pseudoglycal rearrangemen t,40 Glycols, see I ,2-Diols Glycoside, 29 Glyoxal, condensation with phenols, 487 Glyoxylic acids, 496 decarboxylation, 496 Gramacidin A, 404 Gramacidins, 404 Grignard reaction, of acetylenes, 8-19,62 of acetylenic Grignards, 125. 129 of acyloxymethylpyrroles, 162 of aldehydes, 8-19 of chloroformylpyr'roles, 165
577
of coumarins, 469 of halomethylpyrroles, 162 of 0-keto enol ethers, 129 of lactones, 469
of polycarbonyl compounds, 356, 360
of pyrrolic Grignard reagents, 162 in self-condensation of pyrrolic Grignards, 199 of tetrahydropyrones, 429 of apunsaturated ketones, 125 Grignard reagents, using cadmium derivatives, 527 Guttiferae, 534 Hageman's ester, in synthesis of Corrins, 26 2 a-Haloacetaldehyde, condensation with 2aminocrotonate, 151 0-keto esters, 151 a-Haloacetals, hydrogenolysis, 4 1 a-Haloaldehydes, condensation with 2-aminoctotonate, 151 0-keto esters, 151 Haloalkoxylation, with 1.3-dibromo-5,5dimethylhydantoin, 38,41 of olefins, 38,40 of vinyl ethers, 34, 35,40,41 a-Halo amines, cyclisation to aziridines, 456 Halocarbonyl pyrroles, condensation with pyrrole Grignard reagents, 162 &-Haloesters, Claisen condensation, 6 1 Reformatsky reaction, 18 a-Halo ethers, 32,34,369,377,378,392, 394,398,448,491 acetal formation, 32, 34,369, 372, 376378,448,491 as alkylating agents, 548 or-amino ether formation, 393, 394, 398 halogen displacement, 47,51 hydrolysis, 32, 34 Haloethyl lactones, base-catalysed cyclisation, 486 Haloformylpyrroles, coupling with Grignard reagents, 165 Halofurans, 28-58 as carbohydrate precursors, 28-58 Halogenation, of acetals, 43,53 at allylic positions, 72 of amino groups, 98 with N-bromosuccinimide, 72,98,
578
Subject Index
345,550 with 1-butylhypochlorite, 185 of chromans, 550 of coumaranones, 493 of ethers, 49 1 of ketals, 231 of P-keto esters, 124 of methylpyrroketones, 185 of methylpyrroles, 157, 160 of methylpyrrolinones, 214 of nitroalkanes, 265 with phenyltrimethylammonium perbromide, 491 with pyridinium hydrobromide perbromide, 231 of resorcinols, 527-531 with sulphuryl chloride, 157, 185 of tosylates, 382 of a$-unsaturated esters, 345 of a,@-unsaturated?-keto esters, 132 of vinyl ethers, 32, 34,47 of xanthones, 546 Halogena live decarboxylation, of pyrrole carboxylic acids, 157, 159 Halogen displacement, by acetylenic carbanions, 12 in alkylation of bromobenzenes, 265 of alkyl halides, 108, 112 of bromopyrroles, 400 of bromopyrromethenes, 214 in chlorobenzenes, 546 of chlorovinylketones, 125 with diazomethane, 444 of vic-dihalides, 38, 39 of a-halo ethers, 47, 5 1,491 of a-haloketones, 108, 113, 125 of halopyrroles, 143-271 by hydrogenation, 41,70, 368, 378, 546 by hydride ion, 108 on lactone formation, 70 by nitrite ion, 382 by reduction, 108, 112, 113, 125, 351, 444 of vinyl halides, 494, 551 5-Halo-S'-halomethylpyrromethenes, condensation with 5-carboxyl-5'-methylpyrromethenes, 172 with S-methyl pyrromethenes, 179, 180 self-condensation, 170, 172, 179 Halohydrins, cyclisation to epoxides, 9,
10.11 Haloketals. dehydrobrornination, 23 I a-Haloketones, condensation with 2-aminocrotonate, 151 p-keto esters, 151 halogen displacement, 108, 113, 125 phosphorane formation under very mild conditions, 530 reduction, 113, 125 Halolactonisation, with N-bromosuccinimide, 18 of 7-keto acids, 235 of &?unsaturated acids, 265 of?.-unsaturated acids, 18, 71,91, 108 Halomesylates, olefin formation, 403 Halomethylketones, 530 a-Halomethylpyrroles, 157, 160 coupling with Grignard reagents, 162 pyrroles, 162 Halomethylpyrrolinones, 2 14 condensation with pyrroles, 214 5-Halo-5'-methylpyrromethenes,self-condensation, 169, 173, 174 Halopyrans, 28-58 as carbohydrate precursors, 28-58 Halopyrroles, 214 halogen displacement, 159,400 Halopyrromethenes, halogen displacement, 214 Halovinylketones, 125 halogen displacement, 125 photochemical 1,2-~ycloadditions,125 Hantzsch synthesis, of pyrroles, 151 Helmin thosporium siccans, 4 79 IIemi-acetal esters, a-halo ether formation, 378 vinyl ether formation, 493,495 Hemiacetal lactones, 490-49 1,495 cleavage, 490491 from 0-hydroxy ethers, 73 selective reduction, 491,495 Hemiacetals, a-chloroether formation, 369, 44 8 dehydration, 47, 51, 377 hydrolysis, 1-75 from lactones, 23,47,51, 64,92, 377, 491,495 oxidation, 447 protection as and regeneration from benzyl ethers, 493
Subject Index Wittig reaction, 92, 109. 114 Hemoglobin, oxidation, 210 Hexahydroxybenzene, hydrogenation, 68 Hexokinase reaction, 7 Hindered hydride reagents, see Reduction, by hindered hydride reagents Hippuric acid, azlactone formation, 355 Hoesch condensation, of benzyl cyanides, 500,503 of phenols, 500,503 Homologation, of carboxylic acids, 355 Hunig’s base, in methylation of amines, 363 Hydrazides, in reduction of esters, 458 in synthesis of peptides, 409-426 Hydrazine, in specific cleavage of P-benzoylpropionyl esters, 289 Hydrazines. 343 oxidation, 343 1,16-Hydride shift, 244 Hydride transfer, in Cannizzaro reaction, 6 Hydroboration, of aldehydes, 43 of vinyl ethers, 41,49, 52 Hydrogenation, of acetylenes, 8-19,48,62, 96, 106, 107, 122. 125,135 of acrylic pyrroles, 155 of 2-acylcyclohexanones stereospecifically, 453 of azides, 53, 341,403,456 of azo-compounds, 448 of benzofurans, 522 of benzopyrans, 522 of b-bilenes, 195 with Bourguel’s catalyst, 10 of chromenochromones, 501 of coumarins, 488 of cyclic ketones, 378 with dimerisation in Zn-CulAcOH, 9 of 1,3-diones stereospecifically, 453 directing effect of silyl ethers, 133 of a,B,Q’,B’-diunSaturatedketones, 441, 442 of epoxides, 398 of hexahydroxybenzene, 6 8 of hydroxydienones, 125 of 0-ketoester cyanohydrins, 219 of ketones, 51 of ketones in the presence of a#-unsaturated ketones, 131 of lactones, 377 with Lindlar catalyst, 8-19
579
of nitroalkanes, 239 of nitroxides, 51, 52 of nitriles, 60 of oximes, 377 of oximinopyrroles, 157 of oxodipyrromethanes, 220 of porphyrins, 202 of pyridones, 158 of pyrrolinones, 219, 221 of pyrrolylmethylpyrrolinones,2 17, 2 19 over rhodium, 441,522 stereochemical control by polar substituents, 381 with sulphuryl chloride, 157, 185 of tetradehydrocorrins, 232 tetrahydroxy benzoquinone, 68 of a$-unsaturated hemiketals, 505 of a,P-unsaturated-keto esters, 133 of a$-unsaturated nitriles, 452 see also Hydrogenolysis; Reduction; and Reductive cleavage trans-Hydrogenation, of acetylenes, 13 with lithium aluminium hydride, 13 Hydrogen bromide, as demethylating agent, 515 Hydrogen fluoride, as dehydrating agent, 351 Hydrogenolysis, of meso-acetoxyporphyrins, 177,187 with Adam’s catalyst, 25, 159 of alkyl halides, 70, 368, 378 of amides, 369 of aryl chlorides, 351, 355,444 of arylketones, 351 of aryl polyethers, 487 of benzylamines, 25,60, 372 of benzylesters, 117, 155,185,429 of benzylethers, 24, 50, 135, 138,369, 409426,447,448 in sensitive molecules, 493 of bromopyrroles, 400 of carbamic esters, 225, 381, 394,409426 of chlorobenzenes, 546 of diethyldithioacetals, 20 of a-haloacetals, 41 of halopyrroles, 159 of a-hydroxyketones, 129 with lithium in methylamine, 138 of 1,2-oxazines, 27
580
Subject Index
see also Hydrogenation; Reduction; and
Reductive cleavage Hydrohalogenation, of vinyl ethers, 32, 34 Hydrolysis, of amides under mild contitions. 369,372, 318,379,393 of t-butyl esters under mild conditions, 189, 195 of ethers in the presence of lactones, 108, 112 with methanolic ammonia, 400 of tetrahydropyranyl ethers in the presence of 1,3-diols and P-hydroxyketones, 109 Hydroperoxychromenes, 5 17,s 18 rearrangement, 517,518 a-Hydroxyacetals, stereospecific synthesis, 32 Hydroxy acids, Ruff degradation, 22 Hydroxyacyl insertion reaction, 424 1,4-HydroxyaIdehydes, equilibrium with 2,3-dideoxytetrofuranose,28. 29 1,5-Hydroxyaldehydes, equilibrium with 2,3,4-trideoxypentopyranose,28, 29 P-Hydroxyamino acids, 5 9 2-Hydroxybenzophenones, 545 oxidative coupling, 540-545 3-Hydroxychrornans, 479 acid-catalysed rearrangement, 478 4-Hydroxycoumarins, 5 12-525 coupling with quinones, 516-521 Wanslick coupling, 5 16-521 7-Hydroxycoumarins, 470 4e-Hydroxycycloheximide, 435 Hydroxydienones, hydrogenation, 125 stability in base, 125 Hydroxy esters, acetate transfer, 357 0-Hydroxy ethers, oxidation, 73 Hydroxyhalogenation, of vinyl ethers, 32 a-Hydroxyketones, hydrogenolysis, 129 0-Hydroxyketones, protection as and regeneration from methoximes, 133 Hydroxylactones, Ruff degradation, 22 Hydroxylation, of 1,3-dihydroxyxanthones, 547 of ketones, 351,355, 359,363 with osmium tetroxide, 525 with oxygen/CeCI,, 351 /OH; 363 /Pt, 355,359 with persulphate, 547
of phenols, 547 of styrenes, 525 of xanthones, 547 see also Epoxidation; Oxidation cis-Hydroxylation, with lead tetra-acetate, 31 of olefins, 1-75, 97 with osmium tetroxide, 1-75, 97 with performic acid, 12 with potassium permanganate, 8-19, 24, 62.68 stereochemical approach control, 45,47, 68 of ad-unsaturated esters, 54 of vinyl ethers, 31 trans-Hydroxylation, of dienes, 67 of olefins, 8-19, 23,43,64,72 with peroxyacetic acid, 23 with peroxyformic acid, 72 with potassium permanganate, 67 with Prtvost reagent, 6 8 with silver benzoate-iodine, 6 8 of vinyl ethers, 32 see also Epoxides, Epoxidation Hydroxymethylenemethyl ethyl ketone, condensation with oximinones, 152 Hydroxypyrrolidones, 2 19 a-Hydroxytosylates, in synthesis of epoxides, 138 Hypoiodite, as oxidising agent, 516-521 h i d e s , 266,452 cleavage, 266 Imines, acid-catalysed cyclisation, 249 in separation of isomeric amino sugars, 391 stability to ozonolysis, 256 lmino ethers, 234-261 condensation with enarnines, 237,239, 24 3 olefins, 238, 239 ad-unsaturated nitriles, 237, 239, 243 conversion to thiolactams, 241, 249 in synthesis of corrins, 234-245 vinylogous amidines, 234-245 Imino thioethers, addition with azidoketene, 34 1 in synthesis of 0-lactams, 34 I Indoles, condensation with alkyl halides, 249
Subject Index Inosamines, 66-75 Inositols, 66-75 Inososes, 66-75 Interstellar gases, 2 lnversion of configuration, of alcohols, 4, 118,360-363,378,391,400,447 Iodine, in cleavage of epoxides, 402 in specific cleavage of ethylthio esters, 292 Iodine azide, addition with pyrrolines, 456 Iodine/sodium acetate, as dehydrogenating agent, 499 Ion-exchange resins, in epoxide cleavage, 22 Isoamyl nitrite, in specific cleavage of anilidates, 293 lsobutylcarbamic esters, as protecting group for amines, 288 lsocyanates, addition of amines, 380 urea formation, 380 Isoeliptone, 507-509 Isoflavans, 5 12-525 Isoflavones, 5 12 cyclopropane formation, 510 reduction, 523,524 in synthesis of rotenoids, 507-513 Isopropylcarbonic anhydrides, as activating groups for alkylation of carboxylic acids, 351 Isopropylidenes, see Acetonides Isoquinolines, 527 Isotopic labelling, of oxophlorins, 193 of rhodoporphyrins, 192 Isoxazoles, 261-268 basecatalysed cleavage to 0-ketonitriles, 261-268 as protecting group for p-diketones, 261268 reduction to p-enaminoketones, 26 1-268 stability to acid, 261-268 Jatmansinol, acid-catalysed rearrangement, 478 Kanamycins, 367-377 Kasugamycin, sugar moiety, 5 1 Ketalisation, with ethylene glycol/ptosyl acidltrimethyl orthoformate, 265 Ketals, halogenation, 231 selective cleavage, 432 Ketenes, addition to olefins, 91, 112, 113
58 1
as intermediates in synthesis of xanthones, 535439,548, 549 stereochemistry of a-alkylation, 92 p-Ketoacetals, condensation with oximinones, 152 a-Keto acids, 496 decarboxylation, 496 p-Keto acids, decarboxylation, 131 yKeto acids, conversion to methylenepyrrolidones, 263 halolactonisation, 235 p-Ketoaldehydes, condensation with oximinones, 152 Ketoamtdes, condensation and cyclisation, 227 in pyrrolinone synthesis, 227 Ketochlorins, 201 Ketodienol ethers, reduction, 120 a-Keto diesters, 268 reduction, 268 p-Ketoenol ethers, conversion to isoxazoles, 262 formylation, 120 Grignard reaction, 129 a-Ketoepoxides, reductive cleavage, 231 p-Keto ester cyanohydrins, hydrogenation, 219 hydroxypyrrolidone formation, 2 19 a-Keto esters, Diels-Alder reaction, 39,40 p-Keto esters, alkylation, 132 base-catalysed cleavage, 444 as carbanionic reactants, 351 condensation with a-aminoketones, 150 a-chloroacetaldehyde, 151 a-chloroketones, 15 1 oximinones, 155 phenols, 488 Cope condensation, 452 decarboxylation, 351 halogenation, 124 oximinone formation, 150 Pechmann condensation, 488-494 reduction, 265 7-Keto esters, methylenepyrrolidone formation, 235 Keto group, in pyrroketones, unusual prop erties, 164, 178, 185 a-Keto lactones, 496 cleavage, 496 Ketones, 527
582
Subject Index
see also Oxobilanes; Oxomethenes; 0x0phlorins; and Pyrroketones addition with acetylenic carbanions, 446 alkylation, 102-107, 115 via enamines, 453 Baeyer-Villiger reaction, 112, 113, 116 in the presence of olefins, 92, 108 as carbanionic reactants, 351 Diels-Alder reaction, 39,40 epoxide formation, 22 formylation, 262 hydrogenation, 50,s I , 378 in the presence of a$-unsaturated ketones, 131 hydroxylation, 351, 355, 359, 363 Michael reaction, 250 oxidative decarbonylation, 210, 21 1 protection as and regeneration from acetonides, 94-138 dioxolanes, 355 dithianes, 97 oximes, 89 reduction, 4 stereospecific hydrogenation, 378 synthesis from acid chlorides, 20 from amines, 97 from vie-diols, 114 via dithioacetals, 97 by oxidative denitration of nitroalkanes, 385 vinyl ether formation, 444 see also Oxobilanes; Oxomethenes; Oxophlorins; and Pyrroketones 0-Ketonitriles, 232, 261-268,526 condensations, 526 protection as and regeneration from methoxyacrylonitriles, 262 Kieltneyera. 539 Knoevenagel reaction, of formylpyrroles, 155 with malonic acid, 155 Knorr synthesis, of pyrroles, 150, 155 Kolbe reaction, 532 Konigs-Knorr condensation, in synthesis of acetals, 369, 372 Lactams, 438 protection as and regeneration from methoxyenarnines, 251 synthesis by hydrogenation of
pyridones, 158 see also Amides; Piperidones; Pyridones; Pyrrolidones; Pyrrolinones 0-Lactams, 337-348 Lactones, amide formation, 246 in N-acylation of methylenepyrrolidones, 235 Crignard reaction, 469 reduction, 23,47, 51,64,92, 109, 351, 317,488,522,523 in the presence of epoxides, 113 selectively, 491,495 synthesis from hemi-acetals, 447 vinyl ethers, 433 under mild conditions, 429 y-Lactones, synthesis from 7,b-unsaturated carboxylic acids, 18,7 1 6-Lactones, synthesis from yformyl esters, 202; see also Coumarins Lankamycin, sugar moiety, 53 Lead acetate, as acetoxylating agent, 527528,530,532 as dehydrating agent, 363 Lead dioxide, as oxidising agent, 442 Lead tetra-acetate, as oxidising agent, 343, 360,448 Lecanora rupicola, 546 Lecanora straminea, 546 Leucomycin, sugar moiety, 48 Lindlar catalyst, 8-19, 107 Lithium aryl, as carbanionic reactant, 478484 Lomatiol, acid catalysed rearrangement, 478 Maclurin, 544 Macrolide antibiotics, 426-434 Magnamycin, sugar moiety, 48 Magnesia, as condensing agent, 2 Magnesium hydroxide, as condensing agent, 3 Malondialdehyde, as carbanionic reactant, 345 Malonic acid, Knoevenagel reaction, 155 Malonic esters, decarboxylation, 268 Malonic ester syntheses, 16 with acid chlorides, 268 with amino esters, 230 with f.amines, 158 Mammea, 484 Mammea africana, 545
Subject Index Mammea americana, 539, 543 Manganese dioxide, as dehydrogenating agent, 533 as oxidising agent, 73, 217,446, 501, 504, 546 Manganese dioxide/ferricyanide, as dehydre genating agent, 499 Mannich reaction, of azaindoles, 158 McFadyen-Stevens reduction, of esters, 230 Medium and large rings, 404434 Meerwein trialkyloxonium salts, in synthesis of corrins, 234-245 imino ethers, 234-245 vinylogous amidines, 234-245 Melicope simplex, 475 Mercuric oxide, as dehydrogenating agent, 533 Mesobilirubin, incorporation of a-oxymescporphyrinferriheme, 21 2 Mesylates, 392 cleavage with neighbouring group participation, 392 as protecting group for oximes, 253.254 stability to acid, 398 Mesylation, of polyalcohols, 391 “Meta”-bridged cycloalkylpyrroles, 23 1 Metaperiodate, as oxidising agent, 343 Methanolic ammonia, in the hydrolysis of esters and amides, 400 Methoximes, as protecting groups for 0hydroxyketones, 133 Methoxyacrylonitriles, as protecting groups for 8-ketonitriles, 262 p-Methoxybenzyl esters, as protecting group for carboxylic acids, 345,429 N-Methoxycarbonylaziidines,rearrangement, 456 Methoxycarbonylmethylchlorins, 202 carbanionic reactions, 206 Methoxyenamines, as protecting group for lactams, 251 7-Methoxymitosene, 455 3-Methoxypyrroles. 227-232 Methylation, of amines, 355, 359, 363 with dimethylsulphate, 487 of polyphenols, 487,535, 546, 547 of quinones, 487 3-Methylbut-2-enyl alcohol, in C-alkylation of phenols, 474-475 2-Methylbut-3-yn-2-01,in synthesis of
583
chromenes, 469 4-Methylcoumarins, oxidation, 488 Methyl cyanodithioformate, DielSAlder reaction, 27,28 in synthesis of heterocycles, 27, 28 Methyl 2,3-dideoxy-DL2-enopyranosides, 56 Methylenepyrrolidones, 234-245, 263,266 N-acylation, 235 addition with HCN,242, 263 nitroalkanes, 238, 263 condensation with thiolactams, 24 1-261 see also Acyl vinyl amines, Enamides, Vinyl amines Methyl isocyanate, addition with amines, 380 Methylketones, 263 synthesis from acid chlorides, 20 4-Methylpyridines, as carbanionic reactants, 158 5-Methylpyrroketones, chlorination, 185 Methylpyrroles, chlorination, 157, 160 Methylpyrrolinones, bromination, 214 5-Methylpyrromethenes, condensation with 5-brorno-5’-bromomethylpyrrome thenes, 179, 180 demethylation, 21 3 Methyl sorbate, Diels-alder reaction, 25 Michael reaction, of ketones, 250 of nitroalkanes, 238 of a-nitroesters, 357 stereospecific, 446 of a$-unsaturated amides, 250 of a@-unsaturatedesters, 238, 351 of a#-unsaturated ketones, 357,444 of vinylethynylketones, 444 Mineral clay, as condensing agent, 2 Mitomycin C, 435 Mixed anhydrides, in synthesis of peptides, 409-426 Monascorubrin, 531 Monensin, 435 Monomethyl di-O-acetyl-Ltartrate,as intermediate in carbohydrate synthesis, 20 Munduserone, 498, 505 Murraya koenigin, 485 Mycophenolic acid, 472 Myoglobin, oxidation, 210
584
Subject Index
Narbomycin, sugar moiety, 47 Neighbouring group participation, in cleavage of mesylates, 392 in deamination of y-amino acids, 24 in displacement reactions of &substituted esters. 343 in a-hydroxyacetal formation, 32 in selective cleavage of aryl ethers, 515, 521, 524, 537-543 Nencki reaction, in synthesis of xanthones, 535-539, 548 Neotenones, 5 12 Neutramycin, sugar moiety, 53 Nitric acid, as an oxidising agent, 4 Nitriles, as electrophiles, 448, 526 hydrogenation to aldehydes, 6 0 as protecting group for olefins, 234-245 reduction, 232 Nitrite ion, halogen displacement, 382 Nitroalkanes, addition with methylenepyrrolidones, 238, 263 qP-unsaturated aldehydes, 100 a$-unsaturated esters, 268 @-unsaturated ketones, 264 P,y-unsaturated ylids, 101 vinylamines, 238 bromination, 265 as carbanionic reactants, 369, 382, 385 hydrogenation, 239 Michael addition, 238 oxidative denitration to ketones, 385 reduction. 97, 100, 101, 359, 382 Nitroamines, 385 Nitrobenzenes, denitration, 540 reduction, 438 p-Nitrobenzyl esters, as protecting group for carboxylic acids, 345 &-Nitroesters, as carbanionic reactants, 357 p-Nitrophenyl esters, in synthesis of p e p tides, 409-426 p-Nitrophenyl sulphite, in synthesis of pnitrophenyl esters, 409-426 Nitrosation, of ureas, 380 N-Nitrosomethylcarbamylazide, 380 in synthesis ofN-nitrosoureas, 380 N-Nitrosoureas, 380 Nitro sugars, 369 reduction, 369 Nitrosyl chloride, addition with dihydro pyranones, 51, 377
vinyl esters, 5 1, 377 vinylethers, 51,52, 377 vinyl lactones, 51, 377 Nitroxides, hydrogenation, 51, 52 Nucleic acids, condensations, 285-286 phosphorylation, 284-286 protection of functional groups, 286-293 Nucleophilic displacement, in aromatic systems, 542 Nucleoside antibiotics, 387-404 Nucleotidyl transferase, in enzymic synthesis of oligonucleotides, 316 Nystatin, 427 Octaethylporphyrin, oxidation, 201 Oleandomycin, sugar moiety, 47,49 Olefins, addition to give cyclobutanones, 91, 112, 113 addition with alkylsulphenyl chlorides, 38 ammonia, 456 ketenes,91, 112, I13 thiolactams, 241 aminohydroxylation, 47-49,60 aziridine formation, 235,456 as carbohydrate precursors, 1-75 condensation with iminoethers, 238, 239 cyclopropane formation, 102, 1 12 Diels-Alder reaction, 30,236,483 epoxidation, 8-19,26, 38-49, 60 haloalkoxylation, 38,40 hydrogenation, stereochemical control by polar substituents, 381 in the presence of vinyl ethers, 441 cis-hydroxylation. 1-75, 91 trans-hydroxylation, 1-75 oxidation by selenium dioxide, 135, 138 oxidative cleavage, 116,245, 250, 251, 360,493 ozonolysis, 15, 19, 122 protection as and regeneration from nitriles, 234-245 reaction with ethyl azidoformate, 102 cisreduction, 199 fruns-reduction, 199 stereospecific epoxidation, 93, t 13 stereospecific hydration, 9 I synthesis from 1,2-diols, 403 epoxides, 403 N-oxides, 60 in synthesis of vinylogous amidines,
Subject Index
585
238,239
as a protecting group for a-amino acids,
125, 129
Oxidation, of alcohols, 384,385 of aldehydes, 4,8 of ellylic alcohols, 118, 125,231,446 Baeyer-Villiger oxidation, 73,91,108,
cis-Olefins, base catalysed isomerisation, synthesis by Wittig reaction, 92, 109, 114,
383
see also Acetylenes, hydrogenation
trans-Olefins, protection as and regeneration from bicyclo[2.2.1.1 heptenes, 130 synthesis by Wittig reaction, 81-138 Oligonucleotides, chromatography by cellulose TLC, 298 DEAE-cellulose, 294 sephadex gel filtration, 305,308 silica gel, 304 synthesis by enzymes, see Enzymes, in synthesis of oligonucleotides polymerisation, 293-299 via polymer supports, 309-311 via ribooligonucleotides, 3 11-313 stepwise condensation, 300-304 with terminal phosphate, 305-308 tricster approach, 304 Olivetol, 481 Oppenauer oxidation, of benzyl alcohols,
501,504
Orcinol, 537 Orsellinic acid, 428,537 Ortho esters, 517,518 as protecting groups for 1,3,4-triols, 395 Orthoformic ester, see Triethyl orthoformate Orthoformic esters, 219, 222, 226 Osmium tetroxide, as hydroxylating agent,
525
Ostreogrycin, 405 Oxaloyl group, cleavage, 202 Oxalyl chloride, as condensing agent, 495 Oxaly chloride/aluminium chloride, as condensing agent, 488,491 Oxaprostaglandins, 133-138 Oxazines, 25 cleavage, 25,27 hydrogenolysis, 27 Oxazolines, 392,456 as intermediates in cleavage of mesylates,
392
as protecting group for o-aminophenols,
439-440
Oxazolones, 338 cleavage, 338
338
112,113,116
with benzoyl peroxide, 21 1 of benzyl alcohols. 501, 504 of a-bilane-l’,8’-dicarboxylic acids, 185 of bilirubic acid, 213 with bromine, 4, 8 with brominelsulphuryl chloride, 165 with N-bromosuccinimide, 447 with t-butyl hypochlorite, 185 of chalcones, 5 12 with rn-chloroperoxybenzoic acid, 345 of chromanols, 501,504 of cyanoalkanes, 498 of dibenzopyrans, 521 with 2,3-dichloro-S,6-dicyano- 1,4-benzoquinone, 118 of gem-dicyanovinylpyrromethanes,203 of dienones, 124
of S,S’-dimethylpyrroketones,177 of vic-diols, 8, 15,73,97,116, 117 with ferricyanide, 516-521 of flavilium salts, 517,518 of hemiacetals, 447 of hemoglobin, 210 of hydrazines, 343 with hydrogen peroxide, 201,498 of P-hydroxy ethers, 73 with hypoiodite, 516-521 intramolecular, 476
with lead dioxide/lead tetra-acetate, 165 with lead tetra-acetate, 97, 177,343 with manganese dioxide, 73, 217,446,
501,504,546
of 4-methylcoumarins, 488 of myoglobin, 210 with nitric acid, 4 of octaethylporphyrin, 201 of olefins, 135, 138 with osmium tetroxide, 1-75,201 of oxophlorins, 187 with periodate, 8, 15,73, 369 with peroxybenzoic acid, 73 with persulphate, 547 with Pfitzner-Moffat reagent, 73,384,385
586
Subject Index
of phenols, 516-521, 547 of phlorins, 200, 209 of polyalcohols, 4, 368, 369 of porphyrinogens, 199-209 of porphyrins, 210, 211 with potassium iodate, 516-521 of pyrroles, 214 of pyrrolylmethylpyrrolinones, 21 7 of pyrromethanes, 165, 213, 215 with quinones, 476 with selenium dioxide, 135. 138,488 with sulphuryl chloride, 165 with thallium 111, 512 of triols, 369 of xanthones, 547 see also Dehydrogenation; Epoxidation; Hydroxylation; Oxidative cleavage; Oxidative coupling; Oxidative cyclisation; Oxidative ring contraction; and Ozonolysis Oxidative cleavage, of benzaldehydes, 547 with in-chloroperoxybenzoic acid, 433 of dienones selectively, 124 of P-diketones, 265 of vic.diols, 8, 15, 73, 97, 116, 117, 343, 360,448 with lead tetra-acetate, 360, 448 with metaperiodate, 343 of olefins, 116, 117, 122, 245, 250, 251, 360,493 with periodate, 265 of purpurins, 202 with sodium periodate/osmium tetroxide, 493 of a$-unsaturated ketones, 251, 254 of vinyl ethers, 433 of vinylogous hemiketals, 265 see also Oxidation; Ozonolysis Oxidative coupling, of o-aminophenols, 422 of catechol, 5 16-52 I with dichlorodicyanoquinone, 544,552 enzymatic, 544 with ferricyanide, 543 of 2-hydroxybenzophenones, 540-545 of hydroxycoumarins, 516-52 1 with lead dioxide, 442 with manganese dioxide, 546 with pernianganate, 543 of phenols, 44 1,442 photochemically, 544
with potassium ferricyanide, 422, 441 see also Oxidation, Oxidative cyclisation Oxidative cyclisation, of alkylallylquinones, 476 with dichlorodicyanobenzoquinone, 472475 of o-dimethylallylphenoIs, 47 1-475 of 1',8'-dimethyl-o,c-biladienes,183 see also Oxidation; Oxidative coupling Oxidative decarbonylation, of ketones, 210, 21 1 Oxidative decarboxylation, of carboxylic acids, 117 with lead tetra-acetate, 117 see also Ruff degradation Oxidative denitration, of nitroalkanes, 385 with potassium permanganate, 385 Oxidative ring contraction, of chromenes, 517, 518;seealso Oxidation N-Oxides, 60 pyrolysis, 60 Oximes, Beckmann rearrangement, 254 hydrogenation, 377 as protecting group for ketones, 119 protection as and regeneration from mesyl esters, 253, 254 reduction, 368, 382, 384 stability to oxidising conditions, 251 Oximinones, I50 condensation with P-diketones, 152, 157 hydroxymethylenemethylethylketone, 152 8-ketoacetals, 152 P-ketoaldehydes, 152 P-keto esters, 155 reduction, 150 Oximinopyrroles, hydrogenation, 157 b-Oxobilane-l',8'-dicarboxylicacids, condensation and cyclisation with trimethyl orthoformate, 187, 189, 191 a-Oxobilanes, 184-187 reduction, 185 in synthesis of porphyrins, 184-187 b-Oxobilanes, 187-194 cyclisation, 187, 191 in synthesis of porphyrins, 187-194 Oxodipyrromethanes, 220 Oxodipyrrornethenes, 220, 225 hydrogenation, 220 Oxomethenes, 187
Subject Index Oxophlorins, 177, 187-194,209-227 isotopic labelling, 193 oxidative decarbonylation, 210,211 photo-oxidation, 187 stability to borohydride, 178 in synthesis of porphyrins, 187-194 tautomerisation, 193 Oxygenation, see Hydroxylation; Oxidation Oxygen ring compounds, 467-555 . a-Oxymesoporphyrin ferriheme, incorporation into mesobilirubin, 212 Oxyporphyrins, see Oxophlorins Ozonolysis, of enol ethers, 256 of olefins, 15, 19, 122 see also Oxidation; Oxidative cleavage Paal-Knorr method, in pyrrole synthesis, 150 Papillionaceae Leguminosae, 525 Paraformaldehyde, in reductive C-alkylation of pyrroles, 154, 155 LParasorbic Acid, 22 Parmelia formosana, 5 35 Pechmann condensation, of 0-keto esters,
488-494
of phenols, 488494 Penicillins, 337-348 Penicillin, 404 Penicillium rubrum, 532 Penicillium sclerotiorum van Beyma, 526 Pentaalkylpyrroles, 154 Pentachlorophenyl esters, mild hydrolysis,
187 Pentahydroxycyclohexanones,66-75 Peptide antlblotics, 404-426 Perhydrothiazines, 342-348 Periodate, as oxidising agent, 369 Permanganate, as oxidising agent, 543 Persulphate, as hydroxylating agent, 547 Pfitzner-Moffatt reagent, 73, 384,385 Phenolic acids, Nencki reaction, 535-539 in synthesis of xanthones, 535439,548 Phenols, acetoxylation, 527-528,530,532 C-alkylation, 473-475 carboxylation, 528, 532 condensation with aldehydes, 487 chromanones, 506-507 glyoxal, 487 P-keto esters, 488 vinyl halides, 494
587
coupling with arylketones, 518 quinones, 516-521 Duff condensation, 507-509.547 formylation, 507-509,547 Hoesch condensation, 500,503 Nencki reaction, 535-539 oxidation, 516421,547 oxidative coupling, 441,442 Penchmann condensation, 488494 protection as and regeneration from benzyl ethers, 487,493,518, 530 carbonate esters, 441 selective benzylation, 449,487 selective methylation, 487,535,546,547 substitution, 467-555 in synthesis of aryl ethers, 467.555 chromenes, 467485 xanthone formation, 535-539 see also Catechols; Resorcinols o-Phenoxybenzoic acids, xanthone formation, 539 Phenylosazones, 4 hydrolysis, 4 Phenyltrimethylammonium perbromide, as brominating agent, 491 Pheoporphyrins, 192 Phlorins, 200 oxidation, 200,209 Phloroglucinols, 467-555 reactions, 467-555 Phosgene, reaction with pyrroles, 165 Phosphate, protection as and regeneration from anilidates, 293 ethylthio esters, 292 trichloroethyl esters, 291 Phosphodiesterases, in enzymic synthesis of oligonucleotides, 319 Phosphoranes, synthesis under mild conditions, 530 Phosphorous oxychloride, complexation with 2-N,N-dimethylamidopyrroles,
165
pyrromethane amides, 187,189, 191 Phosphorous pentoxide, as condensing agent,
530
Phosphorous tribromide/pyridine. as dehydrating agent, 503 Phosphorylation, with A.T.P., 7 of nucleic acids, 284-286 Phthalimides, as deactivating group for
588
Subject Index
amines, 359 as protecting group for amines, 409-426 Phytoalexins, 516, 524, 525 Pinacol rearrangement, of dihydroxychlorins, 201 Piperidine, in cleavage of anisoles, 551 as condensing agent, 495, 505 Pipcridones, 254 condensation with 1,3-diketones, 254 imino ether formation, 251 see also Amides; Lactams Pivaloyl esters, as protecting group for alcohols, 288 Plexaura homomalla, prostaglandin extraction, 84, 119 Polyenones, from 1,3-diones, 120 Polymerisation, in synthesis of oligonucleotides, 293-299 Polymer supports, in synthesis of oligonucleotides, 309-31 1 Polymixins, 404, 413419 Polynucleotide kinase, in enzymic synthesis of oligonucleotides, 319 Polynucleotide ligase, in enzymic synthesis of oligonucleotides, 320-321 Polynucleotidyl phosphorylase, in enzymic synthesis of oligonucleotides, 317318 Polyphosphoric acid, as condensing agent, 500, 510, 526, 535-539 as dehydrating agent, 355 Porphobilinogen, incorporation of 6aminolevulinic acid, 156 Porphyrinogens, 187, 199-209 oxidation, 199-209 Porphyrins, acetylation, 173 formylation, 182 Friedel-Crafts reaction, 173, 182 hydrogenation, 202 metabolism, 193, 194 oxidation, 210,211 reduction, 199-209 synthesis from a,c-biladienes, 179-184 b-bilenes, 194-199 a-oxobilanes, 184-187 6-oxobilanes and oxophlorins, 187-194 pyrroketones, 176-179 pyrroles, 166-167 pyrromethanes, 167-169 pyrromethenes, 169-176
Potassium carbonate, as condensing agent, 14 Potassium ferricyanide, in oxidative couplings, 422,441 Potassium iodate, as oxidising agent, 516521 Potassium permanganate, in oxidative denitration of nitroalkanes, 385 Prdvost reagent, 68 Pristinamycins, 404 Prodigiosin, 227-232 Prostaglandins, 81-138 biosynthesis and metabolism, 84-89 Prostanoic acid, 82 Pterocarpins, 512-525 Purine Nucleosides, 390-398 Purpurins, 202 oxidative cleavage of isocyclic ring, 202 Pyranocoumarins, 484 Pyran-4-oncs, see Xanthones; lsoflavones Pyranoxanthones, 548-555 Pyrans, as carbohydrate precursors, 28-58 conformational analysis, 28-58 Pyridine, as condensing agent, 552 Pyridinium chloride, as condensing agent, 542 Pyridinium hydrobromide perbromide, as halogenating agent, 231 5’-Pyridinium methylpyrroketones, coupling with pyrromethane-5-carboxylic acids, 185 Pyridones, hydrogenation, 158; see also Amides; Lactams Pyrones, condensation with carbanionic reactants, 526 Pyronoquinones, see Amphipyrones Pyrroketones, 164-165 abnormal keto function, 164 modified nucleophilicity, 177, 185 reduction, 164 in synthesis of porphyrins, 176-179 Pyrrole, polymerisation with aldehydes, 150 Pyrrole carboxylic acids, decarboxylation, 157,159,217,220,222,226,230 iodinative decarboxylation, 157, 159 Pyrrole Grignard reagents, 162 condensation with acyloxymethylpyrroles, 162 alkoxymethylpyrroles, 162 chlorocarbonylpyrroles, 162
Subject Index halomethylpyrroles, 162 self-condensation, 199 Pyrroles, addition with A'-pyrrolines, 230 condensation with acyloxymethylpyrroles, 162 alkoxymethylpyrroles, 162 bromomethylpyrrolinones, 2 14 5,5'-diformylpyrromethanes, 183 dimethylamidopyrrole/POCI,complex, 165 ethyl orthoformate, 219, 222, 226 formaldehyde, 161 formylpyrroles, 143-271 halomethylpyrroles, 162 halomethylpyrrolinones, 2 14 phosgene, 165 pyrrolidones, 230 2-pyrrodinones, 230 formylation with HCN,219, 223 Gatterman reaction, 219 oxidation, 214 reductive C-alkylation, 154-155 synthesis by, formation of C-N bonds, 150,232 3-4 and C-N bonds, 150-152 2-3 and C-N bonds, 152-153 2-3 and 4-5 bonds, 153-154 from other heterocycles, 154 in synthesis of porphyrins, 166-167 Vilsmeier condensation, 230 Pyrrolidones, 217,219,221, 235,239,246 condensation with pyrroles, 230 a$-unsaturated nitriles, 243 imino ether formation, 249 thiolactam formation, 241, 249 Vilsmeier reaction, 230 see also Amides; Lactams Pyrrolines, addition of iodine azide, 456 synthesis from isoxazoles, 26 1-268 A'-Pyrrolines, addition with pyrroles, 230 Pyrrolinones, 214,219,227 as carbanionic reactants, 219 condensation with formylpyrroles, 219, 220,225 hydrogenation, 221 reduction, 219 Vilsmeier condensation, 230 see also Amides; Lactams Pyrrolinylpyrroles, 230 dehydrogenation, 230
589
Pyrrolo( 1,2-a)indoles, 455457 carbomethoxylation, 456 Pyrrolopyrimidine Nucleosides, 398400 N-substitution, 400 Pyrrolylmethylenepyrrolinones,2 17 condensation with formylpyrrolylmethy Ienepyrrolinones, 217 formylation, 217 reduction, 219 Pyrrolylmethylpyrrolinones,2 15 condensation with formylpyrrolylmethylpyrrolinones, 217 dehydrogenation, 217 formylation, 217 hydrogenation, 217, 219 oxidation, 217 Pyrromethane amides, complex formation with POCI,, 187,189, 191 condensation with pyrromethanes, 187, 189, 191 Pyrromethane-a-carbox ylic acids, condensation with a-formylpyrromethanes, 195, 197 5'-pyridiniummethylpyrroketones, 185 Pyrromethanes, 161-163,213 acid-catalysed isomerisation, 167 condensation with 5,5'-di(bromomethyl)pyrromethene, 169 5,5'-diformylpyrroketones, 177 5,5'-diform ylpyrromethanes, 168 formic acid, 167 a-formylpyrromethanes, 195, 214 pyrromethane amides, 187, 189, 191 oxidation, 165,213, 215 in synthesis of porphyrins, 166-167 Pyrromethenes, 163-164,214 condensation with formylpyrromethanes, 214,215 Friedel-Crafts reaction, 180 reduction, 225 in synthesis of porphyrins, 169-176 N-Pyruvylideneglycinatoaquocopper(ll), in Akabori reaction, 60 Quercitols, 66-75 Quinone methides, 471485 cyclisation, 471-485 reduction, 476479 Quinones, addition with dihydrofurans, 487 vinyl ethers, 487
590
Subject Index
benzylation, 487 condensation with chromanones, 5 12 coupling with hydroxycoumarins, 5 16521 coupling with phenols, 516-521 Diels-Alder reaction, 356, 360 as intramolecular oxidising agents, 476 methylation, 487 reduction, 356 ?Radiation, as condensing agent, 2 Radicicol, 427 Reduction, of acetylenes, 13, 14 of acetylenic ethers, 471-472 of acid chlorides, 21, 58 of acylazides under mild conditions, 458 of alcohols, 378 of aldehydes, 2 1, 43 of alkyl halides, 108, 112 with aluminium amalgam, 97, 100, 101, 34 3 of aryl ethers, 115, 251,265 of aryl ketones, 476-479, 533 of azides, 235, 343,376,378,403,456 of bilirubin, 213 Birch reduction, 115 with borohydride, 178,476-479, 501, 502,504,509,523,524 of carboxylic acids, 4, 18 of chlorins. 209 of chromanones, 476,533 of chromenes, 472 of chromenochromones, 501-502 with chromous acetate, 118 of coumarins, 490, 522 of cyclobutylketones, 125 of 2,4-diacetyloxophlorins, 178 with diborane, 43, 164, 522, 533 with di-i-butyl aluminium hydride, 109, 113 of dienediones selectively, 13 1 with disiamylborane. 18, 491, 495 electrolytic, 209 of epoxides, 36.45.46 of a,p-epoxy ketones, 118 of esters, 230 under mild conditions, 458 of a-haloketones, 113, 125 of hemiacetal lactones, 491,495 by hindered hydride reagents, 18, 20, 108,
109,113, 115,439,491.495 of hydrazides under mild conditions, 458 of isoflavones, 523, 524 of isoxazoles, 261-268 of ketodienol ethers, 120 of a-ketodiesters, 268 of 0-ketoesters, 265 of lactones, 23,47,51,64,92, 109, 351, 377,488,491,495,522,533 in the presence of epoxides, 113 with lithium aluminium hydride, 13, 14, 23, 36,4547, 51,54, 56,64 with lithium tri-t-butoxyaluminium hydride, 20, 115 McFadyen-Stevens reduction, 230 of nitriles, 232 of nitroalkanes, 97, 100, 101, 359, 382 of nitrobenzenes, 438 of nitrosugars, 369 of oximes, 368, 382, 384 of oximinones, 150 of a-oxobilanes, 185 of polycarbonyl compounds, 351,491, 495 of polyesters, 133 of porphyrins, 199-209 of pyrroketones, 164 of pyrrolinones, 219 of pyrromethenes, 225 of pyrrolylmethylenepyrrolinones,219 of quinone methides. 476-479 of quinones, 356 Rosenmund reduction, 21,58 with sodium amalgam, 4.21, 368 with sodium borohydride, effect of silylation, 118 with sodium dithionite, 345,403,438 of sulphoxides, 345 in synthesis of cis-fused rings, 502, 504, 509 of L-tartaric acid, 20 of tosylates, 54 with tri-r-butyl stannous hydride, 108 of a$-unsaturated ketones, 56, 501-504, 509 of vinyl bromides, 550 of vinyl ethers, 501. 502.504, 509 see also Hydrogenation; Hydrogenolysis; Reductive cleavage; and Reductive Dimerisation
Subject Index cis-Reduction, with di-imide in pyridine, 199 of olefins, 199 trans-Reduction, of olefins, 199 with sodium in alcohol, 199 Reductive C-alkylation, of pyrroles, 154155 Reductive cleavage, of aryl ethers, 441 of aJ-epoxyketones, 118 with sodium in liquid ammonia, 235 of 1,1,2,2-tetracarboxylic esters, 235 of thioethers, 393 of tosylates, 378 see also Hydrogenation; Hydrogenolysis; Reduction Reductive dimerisation, of acrolein, 9 Reductive methylation, of amines, 381 with HCHOIH, /Pd, 381 Reformatsky reaction, of aldehydes, 18 of a-haloesters, 18 of a,p-unsaturated aldehydes, 18 Resolution of optical isomers, 4, 7,9, 17, 53,99, 108, 129, 135, 138,202, 222,226,245,262,265, 267, 338, 425,441,444,458,503-504,532 Resorcinols, 467-555 acetoxylation, 527-528,530, 532 aldimine formation, 527 alkylation, 528-534 carboxylation, 528, 532 ethanolysls, 531 formylation, 528-534 halogenation, 5 27-53 1 reactions, 467-555 see also Phenols 0-Resorcylic acid, 537 Rhodoporphyrins, isotopic labelling, 192 Ribonucleases, in enzymic synthesis of oligonucleotides, 318-319 Ribooligonucleotides, in synthesis of oligonucleotides, 311-313 Rifamycin B, 435 Ring compounds containing oxygen, 467555 Ring contraction, of chromenes, 517,518 of cyclohexenes, 360 Ringexpansion, by migration to exocyclic carbonium centre, 29 of thiazolidines, 345 RNA Polymerase, in enzymic synthesis of
591
oligonucleotides, 316-31 7 Rosenmund reduction, of acid chlorides, 21, 58 Rotenoids, 498-5 13 Rotenone, 498 Rothemund reaction, in synthesis of porphyrins, 150 Rotiorin, 531 Rotoxen, 498 Rubropunctatin, 53 1 Ruff degradation, of hydroxy acids, 22 of hydroxylactones, 22 Salicylic acids, Nencki reaction, 535-539 xanthone formation, 535-539 Saramycetin, 404 Schiff s bases, 202 as intermediates in porphyrin synthesis, 169 Schmidt degradation, urethane formation, 215 Sclerotiorin, 526, 527 Selective hydrolysis of acetonides, 138 of esters, 133 Selective reduction, of esters, 133 with sodium borohydride. 133 Selenium dioxide, in dehydrogenation, 44 1 as oxidising agent, 488 Selinetin, acid-catalysed rearrangement, 478 Sephadex gel filtration, in chromatography of oligonucleotides, 305, 308 Serratamolide, 406 Serratia marcescens. pyrrole content, 227 1,ZShift. of acyl group in acylvinylamines, 236 Silica gel, as acid catalyst in rearrangement of ally1 ethers, 553-554 in chromatography of oligonucleotides, 304 Silver oxide, as condensing agent, 549 Silyl ethers, directing effect, in borohydride reduction, 1 18 in hydrogenation, 133 as protecting group for alcohols, 134, 136 Sodium bisulphate, as dehydrating agent, 476 Sodium dithionite, as reducing agent, 345, 403,438 Sodium periodate/osmium tetroxide, as oxidising agent, 493
592
Subject Index
Sorbus aucuparia, extraction of L-parasorbic acid, 22 Spiramycin, sugar moiety, 48, 52 Spirographis spallanrani, porphyrin moiety of heme, 189 Sporidesmolide-I, 406 Stabilised carbanions, 59; see also Acetylenic carbanions; Aldol condensations; Carbanionic Reactants; Crignard reactions; Malonic ester syntheses Staphilomycins, 404 Stereospecific hydration, of olefins, 91 Sterigmatocystin, 485 Strecker reaction, 338 Streptomyces antibioticus, 4 19 Streptonigrin, 435 Streptovitacin A, 435 Styrenes, hydroxylation. 525 &Substituted esters, displacement reaction, 34 3 Sugars, see Carbohydrates Sulphide contraction, in inter- and intramolecular reactions, 241-261 Sulphoxides, 345 reduction, 345 in ring expansion of thiazolidines, 345 Sulphuryl chloride, as halogenating agent, 157, 185 as oxidising agent, 165 Swartzia madagascariensis, 52 1 Symphonia globulifera, 544 L-Tartaric acid, as carbohydrate precursor, 19-24 reduction, 20 I , 1,2,2-Tetracarboxylic esters, reductive cleavage, 235 Tetracyclines, 348-364 Tetradehydrocorrins, hydrogenation, 232 2-Tetrahydrofuranyl ethers, 31 Tetrahydrofurobenzofurans,488 Tetrahydropyranyl ethers, as protecting group for alcohols, 96-138, 291 Tetrahydropyrones, 429 Grignard reaction, 429 Tetrahydroxybenzoquinone,hydrogenation, 68 Tetramethylammonium hydroxide, as condensing agent, 541 Thallium 111, as oxidising agent, 512
Thallium hydroxide, as condensing agent, 5 Thiazines, 342-348 isomerisation, 345 Thiazoles, addition with azidoketene, 341 in synthesis of penicillins, 34 1 Thiazolidines, 337-343 cleavage, 345 as protecting group for a-aminothiols, 342-345 ring expansion, 345 substitution, 343 Thiazolines, addition with azidoketene, 341 Thiazolones, 360 cleavage, 363 as protecting group for aldehydes, 362 Thioacetals, 382 Thioamides, hydrolysis, 363 methylation, 363 as protecting group for amines, 363 Thiobenzoyl glycine, thiazolone formation, 360 Thioethers, as carbanionic reactants, 343 cleavage, 393 Thioformylpyrromethanes, 202 reactivity, 202 Schiffs base formation, 202 1-Thioglycosides, 27,28 Thiohemiacetals, 343 Thioimino ethers, hydrolysis, 363 as intermediates in thioamide hydrolysis, 363 Thiolactams, 24 1-26 1 addition with olefins, 241 condensation with methylenepyrrolidones, 241-261 in synthesis of corrins, 241-261 vinylogous amidines, 241-261 Thiolactones, 257 ring cleavage to amides, 258 Thiols, 338,347 Titanium tetrachloride, as catalyst for Fries rearrangement, 442 Tosyl amides. as protecting groups for amines, 409-426 Tosylates, displacement by azide ion, 343, 378 halogenation, 382 reductive cleavage, 54, 378 Tosylation, of polyalcohols, 376, 378, 382, 395
Subject Index Tosyl chloride/collidine, as dehydrating agent, 478 N-Tosylglycines, condensation with a$unsaturated ketones, 153 Trialkyl oxonium salts, in synthesis of corrins, 234-245 imino ethers, 234-245 vinylogous amidines, 234-245 Trichloroethyl esters, as protecting group for carboxylic acids, 89, 345 phosphate, 291 Trichloromethylsulphenylchloride, addition to vinyl ethers, 34, 35 Tricyclooctanes, 47 1 2,3,4,-Trideoxypentopyranose,equilibrium with l,S-hydroxyaldehydes, 28, 29 1,3,5,-Trienes, cyclisation, 202 Triester approach, in synthesis of oligonucleotides, 304 Triethylammonium acetate, as dehydrating agent, 360 Triethyi orthoformate, in synthesis of acetals, 487 Trifluoroacetic anhydride, as condensing agent, 532 Trifolirhizin, 524 2',4',7-Trimethoxyisoflavone, 5 2 1 Trimethyl orthoformate, condensation with b-bilene-l',8'-dicarboxylicacid, 185 b-bilenes, 195, 197 b-oxobilane-1',8'-dicarboxylicacids, 187, 189,191 1,3,4,-Triols, protection as and regeneration from orthoesters, 395 Trioxymethylene, condensation, 2 Tripyrrylmethanes, as interrnidiates in pyrromethane synthesis, 168 side-products from b-bilene cyclisation, 197 Tritylamines, as protecting groups for amines, 341,347,409426 Trityi ethers, as protecting group for alcohols, 400, 402 Tubaic acid, 505, SO6 Tubanol, 503 Tyrocidins, 404 Tyrothricin, 406 Ubichromenol, 476 Ubiquinone, rearrangement to
593
ubichromenol, 476 Ullman synthesis, of aryl ethers, 539 Umbelliferone, 470 Umtatin, 512 a,P-Unsaturated acids, addition of amines, 35 1 P,y-Unsaturated acids, alkylation, 267 bromolactonisation, 265 ?-Unsaturated acids, cyclopentenone formation, 488,490-494 halolactonisation, 18,71, 91, 108 a,P-Unsaturated aldehydes, 345 1,2-addition, 480 1,rl-addition of amides, 345 addition with carbanionic reactants, 100 Diels-Alder reaction, 30,4 1, 43 reaction with alkylating agents, 1 1 3 Reformatsky reaction, 18 a,@-Unsaturatedamides, 250 Michael reaction, 250 see also Pyridones; Pyrrolinones Unsaturated, deoxy, I-thiosugars, 27, 28 qb-Unsaturated esters, 235, 239, 345 addition of HCN, 268 nitroalkanes, 268 condensation with a-amino esters, 230 decarboxylation, 235, 239 Diels-Alder reaction, 30, 70-72, 236 halogenation, 345 cis-hydroxylation, 54 Michael reaction, 238, 351 from pyrolysis of 0-acetoxyesters, 71 P,T-Unsaturated esters, 439 a$-Unsaturated hemi-ketals, 505 hydrogenation, 505 a,P-Unsaturatedy-keto esters, Diels-Alder reaction, 245 halogenation, 132 hydrogenation, 133 a$-Unsaturated-keto-ketones, base catalysed cleavage, 444 a,P-Unsaturated ketones, addition with HCN, 124,232,262 nitroalkanes, 264 condensation with N-tosylglycines, 153 cyclopropane formation, 509 dehydrogenation, 441 Diels-Alder reaction, 483 epoxidation, 231 with Crignard reagents, 125
594
Subject Index
Michael addition, 357,444 oxidative cleavage, 251, 254 photochemical 1,2-~ycloaddition,125 reduction, 56, 501-504, 509 reductive rearrangement to allylic alcohols, 23 1 a,p,yUnsaturated ketones, 47 1-475.480 485 cyclisation to chromenes, 471-475,48& 485 Diels-Alder reaction, 554 a$-Unsaturated nitriles, 235, 239,439,452, 526 condensation with iminoethers, 237, 239, 24 3 pyrrolidones, 244 Diels-Alder reaction, 30, 93, I08 hydrogenation, 452 a$-Unsaturated nitroalkanes, 385 addition of ammonia, 385 Diels-Alder reaction, 97 0,r-Unsaturated ylids, addition of nitroalkanes, 101 Ureas, 380 nitrosation, 380 Urethanes, as protective precursors of vinyl groups, 2 15 from Schmidt degradation, 215 Valinomycin, 406 Verdohemes, 2 11 Vilsmeier condensation, of pyrroles, 230 of pyrrolidones, 230 of 2-pyrrolinones, 230 in synthesis of bipyrroles, 230 Vilsmeier-Haack procedure, formy lation of pyrrolylmethylenepyrrolinones,2 1 7 pyrrolylmethylpyrrolinones,21 7 Vinyl amines, 230,234,261,439 addition with nitroalkanes, 238 cyclisation to thiazines, 347 in synthesis of vinylogous amidines, 234245 Vinylation, of aromatic compounds, 476 Vinylcoumaranones, 509-513 rearrangement, 509-51 3 Vinylene carbonate, Diels-Alder reaction, 67,68 Vinyl esters, Diels-Alder reaction, 70, 71 NOCI addition, 51, 377
Vinyl ethers, 377,395,429,433,444,467555 from acetals, 41 acid-catalysed cyclisation, 43 addition with NOCI, 51, 52, 377 quinones, 487 sulphuryl chlorides, 34, 35 alcoholysis, 429 amination, 52 aminoalkoxylation, 52 cleavage, 230 condensation with a-amino esters, 230 Diels-Alder reaction, 30, 41 displacement of alkoxide by amines, 230 epoxidation, 31 haloalkoxylation, 34, 35, 40, 41 halogenation, 32,34,48 from hemiacetals, 47, 5 1 hydroboration, 41,49, 52 hydrohalogenation, 32, 34 hydroxyhalogenation, 32 cis-hydroxylation, 3 1 trans-hydroxylation, 32 oxidative cleavage, 433 ozonolysis, 256 reduction, 501, 502, 504,509 in synthesis of thiazolidines, 339 Vinylethynylketones, 446 double Michael addition, 445 Vinyl group, protection as and regeneration from urethanes, 215 Vinyl halides, condensation with phenols, 494 reduction, 550 in vinylation of aromatic compounds, 476 Vinyl lactones, addition of NOCI,51, 377 Vinylogous amidines, 234-26 1 in synthesis of corrins, 234-261 Vinylogous hemi-ketals, oxidative cleavage, 265 Vinyl thio-ethers, 345 Vinyltriphenylphosphonium bromide, 456 Viomycin, 404 Vitamin B,,, 232-268 Wanslick coupling, of catechols, 516-521 . of hydroxycoumarins, 516-521 Wessely-Moser rearrangement, of xanthones, 535 Wittig reaction, of a-aminoketones, 439
Subject Index
of enolates, 120 of hemiacetals, 92, 109, 114 in synthesis of cis-olefins, 92, 109, 114, 383 fruns-olefins, 81-1 38 &r-unsaturated esters, 439 vinylamines, 439 under very mild conditions, 530 Wolff rearrangement, of diazoketones, 20 Woodward’s reagent, in synthesis of p e p tides, 422
Xanthones, 495-5 13,534-555 cleavage, 499 formylation, 547
595
halogenation, 546 hydroxylation, 547 oxidation, 547 substitution, 534-555 Wessely-Moser rearrangement, 535 Xanthyletin, 478 Xanthyllum salts, as intermediates in synthesis of xanthones, 546 Zinc carbonate, as condensing agent, 494 Zinc chloride/phosphoryl chloride as condensing agent, 535-9,548 Zinc/Copper, in specific cleavage of trichloroethyl esters. 291
Total Synthesis OfNatural Products, Volume1 Edited by John Apsimon Copyright © 1973 by John Wiley & Sons, Inc.
Reaction Index 3-Acetamido-2,3dideoxy-D-tetrose, 58 3-Acetamido-2,4,6-tri4-benzyl-3deoxya-Dglucopyranosyl chloride, 369 a-Acetoxy-protoporphyrin-IX dimethyl ester, 178 I-Acetylbchloroquinide, 70 4-Acetyla-hydroxydeuteroporphyrin-IX, 178
N-Acetyl lincosamine, 385 Acronycine, 476,484 Acronylin, 475 a-Acrosazone, 4 a-Acrose, 4 a-Acrosone, 4 Actinomycin D,419422 Aetioporphyrins, see etioporphyrins Aflatoxin B, ,486,490,491 Aflatoxin B, ,486,488 Aflatoxin G , ,494 Aflatoxin M,,486,491494 Ageratochromene, 469,471,476477 Alanine t-RNA gene, 321-324 Alleovodionol, 475 methyl ether, 473474 Allitol, 9, 10 DL-AUose, 72 up-DL-Allose, 45 Alvaxanthone, 553 7-Aminocephalosporanic acid, 345 5-Amino-5deoxya-Dalofuranuronic acid, 400 2-Amino-2deoxy4,54-isopropylidene-
D-pentanoic acid, 60 5-AminoJ,6dideoxy-DLallonic acid, 25 5-Amino-5,6dideoxy-DL-gulonic acid, 27 2-Amino-3,3dimethyl4-hydroxybutyraldehyde, 60 DLsrythro-4 -Amino-5-hydroxyhexanoic acid, 51,377 2-Aminomethyl-3,5-bis(carboxylmethyl)pyrrole, 159 2-Aminomethyl-3carboxyl methyl pyrrole, 159 6-Aminopenicillanic acid, 341
4-Amino-2,3,4-trideoxya-D~rythrohe~-
2-enpyranosiduronic acid, 403 erythro-2-Amino-l,3,4-trihydroxybutane, 19 Ampicillin, 338 Angustmycin A, 388,394 1,4-Anhydro-cis-conduritol carbonate, 67 1,6-Anhydro4deoxy6-DL-ribo-hexopyranose, 46 1,6-Anhydr0-3deoxyS-DL-ribo-hexopyranose, 45 1,6-Anhydro4deoxyQ-DL-xy/o.hexopyranose, 4 3 1,6-Anhydro4deoxy-3-O-methylQ-DLxylo-hexopyranose, 54 1,6-Anhydro3,4dideoxyQ-DL-erytbrohex3enopyranose, 43 1,6-Anhydro-3,4dideoxyQ-DL-erythro597
598
Reaction Index
hexopyranose, 46 1,6-AnhydroQ-DL-glitco-hexopyranose, 45 1,6-Anhydro-3C)-met hylQ-DL-ghco-hexopyranose, 45 Anthramycin, 434,437440 D-Apionic acid, 8 Apiose, 6 L-Apiose, 22 Aposclerotiorin, 530 DL-Arabinitol, 11 DL-Arabinose,6, 13,14 Arthothelin, 546 Ascochitine, 528 Aversin, 496 methyl ether, 486 Beauvericin, 422424 BeUadifolin, 547 5~-Benzyl-2,3dideoxy-D-pentofuranose, 24 DQ-Benzyloxydecanoic acid, 425 a-Benzyloxyprotoporphyrin-IXdimethyl ester, 178 truns-3-BenzyIthio-2chlorotetrahydropyran, 35 5-Benzylthio-3,4dihydro-2H-pyran, 34, 35 lruns-3-Benzylthio-2-rnethoxytetrahydropyran, 34,35 Bilirubic acid, 21 3 Bilirubin, 213,215 Biliverdin-IXa, 210,212,213,215 Biliverdin dimethyl ester, 179
N,N’-Bis(carbobenzyloxy)-2deoxystreptamine, 369 N,N-Bisdemethylterramycin, 363 Blasticidin S, 404 3-BrOmO-S,6dihydro4H-pyran, 3 2 3a-Bromo-2a,4pdknethoxytetrahydropyran, 38 3cr-Bromo-2a,Sadime t hoxytetrahydropyran, 42
3P-Bromo-2a,4adirnethoxytetrahydropyran, 38
~runs-&cis-4-Bromo~,8dioxabicyclo [ 3.2.1 ] octane, 43 3-Bromo-2ethoxy-5.6dihydro-ZH-pyran, 39 Buchanaxanthone, 551,552
2-t-Butoxy-3,4epoxytetrahydropyran, 36 2-1-Butoxy-3-hydroxytetrahydropyran, 36 But yl cis4-acetoxy-5,6 dihydro4 H-pyran 6-carboxylic acid, 40 Butyl trans-2-acetoxy-5,6dihydro-2H-pyran 6carboxylic acid, 40 Butyl truns4-acetoxyJ,6dihydro4H-pyran 6carboxytic acid, 40 Qnnabichromene, 473474,481 Cannabicyclol, 483
N-CarbobenzyloxydesformylgramicidinA,
409 N-Carbobenzyloxypurornycin,394 5-Carboethoxy4ethoxyacetyl-3ethoxypropionyl-2-methylpyrrole, 155 2,30-CarbonylnoviosyI chloride, 448 3-Carboxyethyl-2,5dimethylpyrrole.150 5-Carboxyethyl-2,3dimethylpyrrole,153 2-Carboxyethyl-5-rnethylpyrrole,152 6-(Carboxymethyl)acetylrhodoporphyrinXV dimethyl ester, 191 Cephaloridine, 345 Cephalosporins, 342-348 Cephalosporin C, 345 Cephalosporin-V, 345 Cephalothin, 345 DL-Chalcose, 53 Chlorine, ,202 Chlorin-e,, 169,202 Chloroamphenicol, 434,457458 Chlorocruoroporphyrin, 182 3-CNoro-5,6dihydro4H-pyran, 32 2-@-Chloroethoxy)-3-hydroxytetrahydropyran, 3 1
4p-Chloro-3aethylthio-2p-methoxytetrahydropyran, 38
Iruns-2-Chloro-3ethylthiotetrahydropyran,
35 truns-2-Chloro-3-methylthiotetrahydropyran, 35 DL-Cinerulose A, 56 Citrinin, 532 Citrylidenecannabis, 483 Clausenin, 476477 Cobyric acid, 270 Conduritol-C, 67 Coproporphyrin-I, 170, 174
Reaction Index Coproporphyrin-11, 167,169, 174
Coproporphyrin-111,174,187 tetramethyl ester, 187
Coproporphyrin-IV, 174,187 Cordycepin, 387,398 a-Corrnorsterone, 254 0-Corrnorsterone, 254 Coumestrol, 515,518 Coumurrayin, 472 Curvularin, 427429 Cycloheximide, 434,451454 Cycloserine, 434
Deguelin, 501 Dehydrodeguelin, 501 Dehydroelliptone, 509 15-Dehydroprostaglandin E, , 122-124 Dehydrorotenone, 505-506 Demethylsuberosin, 47 2 D-Dendroketose, 7 LDendroketose, 7 12a-Deoxy-5a,b-anhydrotetracycline,356 5-Deoxy-DL-arabinonic acid, 23 6-Deoxy-6demethyltetracycline, 349,35 1, 353,359 3-Deoxyepiinositol, 368 2-Deoxy-2-fluoro-DLerythritol,6 1 2-Deoxy-2-fluoro-DL-ribitol,6 1 2-Deoxy-2-fluoro-DL-threitoI,6 1 2-Deoxya,O-DLribo-hexopyranose, 46 3-Deoxy-DLribo-hexopyranose,45 4-Deoxya&DL-ribo-hexopyranose, 46 6-Deoxy-D~robino-hexulose, 64 6-Deoxy-L-xy/o-hexulose, 64 2-Deoxy-3,44l-isopropylidene-DLallitol, 73 Deoxyjacareubm, 552 5-methyl ether, 552 3-Deoxy-2-keto-D-pentaric acid, 64 1-Deoxy-DL-lyxitol, 23 2-Deoxy-D+rythro-pentose, 15 2-Deoxy-L-erythro-pentose,17 2-Deoxy-DL-erythro-pentose,16-18 4-Deoxy-DLerythro-pentose, 47 2-Deoxy-DLthreo-pentose, 17 1-Deoxy-DLribitol, 23 5-Deoxy-DLribonic acid, 23 2-Deoxy-D-ribose, 15 4-Deoxy-DL-ribose, 46,47,73 2-Deoxystreptamine, 368
599
3-Deoxy-2-tetrulosonicacid, 64 2-Deoxy-D-xylose, 14,17 Derrisic acid, 503 DL-Desosamine, 47 Desoxophylloerythroetioporphyrin,197, 198 Deuteroporphyrin-IX, 170 cis-3,6-Diacetoxycyclohexene4-carboxylate, 7 1
2,4-DiacetylaacetoxydeuteroporphyrinIX,178
Di4-acetyl-2deoxy-Lrhreo-pentaric acid dimethyl ester, 20 Di4-acetyl tartaric anhydride, 20 1,6 :2,3-Dianhydro4deoxyQ-D Llyxohexopyranose, 43 1,6:2,3-Dianhydro4deoxyQ-DLribohexopyranose, 43,46
5,6-Dianhydro-2ethoxyb-rnethyl-2H-pyran,
48 1,6 :3,4-DianhydroQ-DLallo-hexopyranose, 43 3a,4~-Dibromo-2aethoxytetrahydropyran, 38
3a,4P-Dibromo-2Pethoxytetrahydropyran, 38
3,5-Dicarboxyethyl-2,4dirnethylpyrrole,
150 6,7-Di(carboxymethyl)methyI-2,3dimethyl1,4,5,8-tetra(carboxymethyI)ethylporphyrin, 169 5,6-Dideoxy-DL-ribo.hexitol, 54 4,6-Dideoxy-Lribo.hexonicacid, 1,5-lactone, 22 5,6-Dideoxy-D~ibo-hexonic acid lactone, 54 3,4-Dideoxya$-DL.-erythro-hexopyranose, 46 2,3-Dideoxya,O-DL-eryrhro-hexose,4 6 5,6-Dideoxy-D-threo-hexulose, 64 4,6-Dideoxy-34l-methyl-D-xylo-hexose, 54 3,5-Dideoxy-L-erythm-pentose,22 2,3-Dideoxy-DL-pentose, 18 cis-2,S-Diethoxy-5,6 dihydro-2H-pyran, 39 rrans.2,5-Diethoxy-5 6dihydro-2H-pyra11, 39 trons-5,6-Diethoxy-5,6dihydro-2H-pyran, 39 Dihydrocitrinin, 533
600
Reaction Index
2,3-Dihydrofuran, 29 Dihydroisoneobilirubic acid, 222 Dihydro4-methylsterigmatocystin,495 Dihydroneobilirubic acid, 219 7,8-Dihydro7-phyUoporphyrinXV, 199 d,l-13,14-DihydroprostaglandinE , , 118-122 3,4-Dihydro-2H-pyran, 29 1,3-Dihydroxy-7-methoxyxanthone, 53 5 1,3-Dihydroxypropanone, 5
trimethyl ether, 475 Flindersine, 480 Formose, 3
2-Formyl-3,4dihydro-2H-pyran,30 2-Formyl4-vinyldeuteroporphyrin-IXdi-
methyl ester, 189 DL-Forosamine, 52 Franklinone, 475 D-Fructose, 4,64 cis-2,5-Dimethoxy-5,6dihydro-2H-pyran, L-Fructose, 4 42,46 Fuscin, 532
cis-2,5-Dimethoxyd-methyltetrahydropyran, 41 trans-2,5-Dimethoxyd-methyltetrahydropyran, 4 1 2,4-Dimethoxytetrahydropyran, 36 cis-2,5-Dimethoxytetrahydropyran,4 1 truns-2,5-Dimethoxytetrahydropyran,4 1 PJ3-Dimethylmethylenebutyrolactam,235 6,8-Dioxabicyclo [3.2.1] octane, 43 6,8-Dioxabicyclo 13.2.1 1 oct-2-ene. 43 6,8-Dioxabicyclo [3.2.1] oct-3ene, 43 Divinylglycol, 9 2.4-Divinylrhodoporphyrin-XVdimethyl ester, 189 Elliptic acid, 509 DL-Epimycarose, 62.64 6-Epipenicillin-V methyl ester, 342 3,4-Epoxy-2-methoxytetrahydropyran,36 Erosnin, 521 Erythritol, 17 D-Erythrose, 19 Ethyl 4-amino4deoxya-DL-lyxopyranoside, 60 4-Ethyl-2cyclododecenol, 23 1 5-Ethylthio-3,4dihydro-2H-pyran, 34.35 trans-3-Ethylt hio-2-met hoxy te t rahydropyran, 34,35 Etioporphyrin-I, 170, 173, 179 Etioporphyrin-II, 167, 169,173 Etioporphyrin-Ill, 173 Etioporphyrin-IV, 173 Euxanthone, 538 Evodionol, 475 methyl ether, 471,484 Flcmingin A, 483 Flemingin B, 483 Flemingin C, 483
Galactitol, 10 a#-DLGalactose, 45 Gentisein. 543 Girinimbine, 485 DGluconic acid, 4 D-Glucose, 4 DLClucosc, 43 DLGlucosone, 4 Glutarimide-3-acetic acid, 451 Gramicidin S,4 0 9 4 12 Griseofulvic acid, 444 Griseofulvin, 434,440447 epi-Griscofulvin, 444 Griseoxanthone C, 546 Harderoporphyrin, 192 Hemin, 170 Hesperimine, 256 D-clrabino-Hexosephenylosazone, 4 DLarabino-Hexose phenylosazone, 4 D-arabino-Hexosulose, 4 DLarubino-Hexosulose, 4 D-arubino-Hexulose, 3 D-xylo-Hexulose, 3 I-Hydroxy-5-bromo-3-oxabicyclo [ 3.2.1 ] octan-2-one. 71 7-Hydroxy-l l ,I 2dimethoxycoumestan, 519 3-Hydroxy-8,9dimethoxypterocarpan, 522 3-Hydroxy-2-methoxytetrahydropyran, 36
cis. & trons-S-Hydroxy-2-methoxytetra-
hydropyran, 41 2-Hydroxymethyl-3,4dihydro-2H-pyran, 43 Hydroxymethylglyceraldehyde, 6 Hydroxymethylglycerol, 6
Reaction Index
4C-HydroxymethyI-Dglycero-pentulose, 7
1-Hydroxy-2,3,4,7-tetramethoxyxanthone, 54 1 Ileu-Gramacidin A, 407409 Inermin, 524 do-lnositol, 68 epi-lnositol, 68 myo-lnositol, 68 neo-lnositol, 68 Isobilirubic acid, 213 Isocannabichromene, 48 1 a-epi-Isocycloheximide, 45 3 Isoelliptone, 509 Isogriseofulvin, 444 lsoneoxanthobilirubic acid, 2 19 Isonovobiocin, 448 Isopemptoporphyrin methyl ester, 187 Isorotenone, 51 1 Isoxanthobilirubic acid, 213
601
Medicagol, 518,519 Mesobilirhodin, 215 Mesobilirubin-IXa, 194,214 Mesobilirubin-XIlb, 214 Mesobiliverdin-IXa, 212 Mesobiliverdin-IW, 212 Mesobiliviolin, 215, 218 Mesoporphyrins, 173 Mesoporphyrin-IV dimethyl ester, 195 Mesoporphyrin-IX, 181 dimethyl ester, 185,187 Mesoporphyrin-X dimethyl ester, 195 Mesoporphyrin-XI dimethyl ester, 187 Mesoporphyrin-XI11dimethyl ester, 195 Metacycloprodigiosin, 231 Methicillin, 338 Methose, 3 2-Methoxy-3,4dihydro-ZH-pyran, 41 2-MethoxyJ,6dihydro-ZH-pyran, 46 3-Methoxy-3,4dihydro-2H-pyran, 41 7-Methoxy-2,2dimethyIchromene,47 1 8-Methoxyhomopterocarpin,5 24
Jacareubin, 484,549
trans-2-Methoxy-6-methoxymethyl-5,6-
Kanamycin A, 372 Kasugamine, 51 Kasugamycin, 377 Kasuganobiosamine, 53,378 Keto-D-psicose pentaacetate, 394
trans-2-Methoxy-3-methylthio-tetrahydro-
dihydro-2H-pyran,41
Lankavose, 53 Lapachenole, 468,469 ' Lichexanthone, 535 Lincomycin, 3 8 1 cis-lincomycin, 381 Lonchocarpine, 470,484 Lucernol, 515 Luvungetin, 470,471 DL-Lyxose, 13 Maackiain, 521-522 new-(T)-Magnesium protoporphyrin-IX, 192 Mahanimbine, 483 Mangiferin, 548 DL-Mannitol, 4,9 DMannonic acid, 4 LMannonic acid, 4 D-Mannose, 4 L-Mannose, 4
pyran, 34,35
4-Methoxypterocarpin,522 3-Methoxypyrrole-2-carboxylicacid, 230 truns.2-Methoxy-3-(trichloromethylthio)-
tetrahydrofuran, 34 Methyl(~)-3-O-acety14,50-isopropylideneshikimate, 71 Methyl 2,3-anhydro4deoxyd-O-methyla-D L-lyxo-hexopyranoside, 40 Methyl 2,3-anhydro4deoxyd0-methyla-DLribo-hexopyranoside, 40 Methyl 50-benzyl lyxfuranoside, 24 Methyl 5-O-benzylQ-D-ribofuranoside, 24 Methyl Dkymaroside, 49 Methyl 4deoxy-DL-hexopyranoside, 40 a-,& p-Methyl4deoxy-30-methyl-DL xylo-hcxopyranoside, 54 Methyl 4deoxy-30-methyla-DLthreopentopyranoside, 36 Methyl 4deoxy-30-methyl$-DLthreopentopyranoside, 36 Me thy1 4deoxyQ-Dhrythro-pentopyranoside, 46 Methyl 3deoxyQ-D-ribofuranoside, 398 des-Methyl deuteroporphyrin IX, 174
602
Reaction Index
Methyl @-2$-5diacetoxya-3,a4-dihydroxycyclohexylcarboxylate, 71 Methyl 2,3di43-acetyi-Sdeoxy-Lthreo4-pentulosonate, 20 Methyl 2,3di43-acetyl-543-ethyl-Lthreo4-pentulosonate, 20 Methyl 2,3di.O-acetyl-L-theuronate, 20 a- & &Methyl 4,6dideoxy-343-methyl-20-p-tolylsulphonyl-DL-xylo-hexopyranoside, 54 Methylenitan, 2 Methyl-Dhpimycaroside, 62 34-Methyla,r3-DL-glucopyranose,45 N-Methylglucosamine, 365 34l-Methyl-DLglucose, 45 Methyl 2,343-isopropylideneQ-DL-allofuranoside, 73 Methyl DL-kasugaminide, 51 443-Methyl-DL-lyxose, 46
Neodulin, 524 Neoisoxanthobilirubic acid, 21 3 Neoxanthobilirubic acid, 213,214,219 p-Nitrophenylserine methyl ester, 457 Norlichexanthone, 535 Noviose, 448 Novobiocin, 434.447450 Octaethylchlorin, 199 Octaethyl-7,8dihydroxychlorin,210 Osajaxanthone, 548 Osthenol, 472 4-Oximino-5axohexanoic acid, 51,377 Oxytetracycline, 359
Pemptoporphyrin, I82 dimethyl ester, 189 D-Penicillamine, 338 Penicillins, 337-348 cis-3-Methyld-methoxycarbonyl-3,6dihy- Penicillin V, 341 dro-l,2axazine hydrochloride, 25 D-threo-Pentulose, 64 Methyl 443-methylaDL-arabinopyranoPheophorbidea, 206 side, 42 7-Phyllochlorin-XV, 199 Methyl 343-methyla$-DLglucopyranoPhylloporphyrins, 176 side, 45 Phylloporphyrin-XV, 173, 181 Methyl 4-O-methyla-DL-lyxopyranoside, DL-Picrocin, 47 41 Pisatin, 524,525 Methyl DLmycaminoside, 48,49 Polygalaxanthone-B, 547 Methyl DL-mycaroside, 62 Polymixin 9, ,416 Methyl DL-oleandroside, 49 Pongachromene, 470 Methyl (W)-shikimate, 72 Porphobilinogen, 155-158 Porphyrina, 169 0-Methylsterigmatocystin, 495 Preremirol, 475 5-Methylthio-3,4dihydro-2H-pyran. 34, Prodigiosin, 227 35 Methyl a-thiolincosaminide, 382 Prostaglandins, 81-138 Methyl 2,3,5-tri-O-acetyl4.4'-anhydro4Prostaglandin E, 89-118 C-hydroxymethyl-L-threo-pentonate, d,f-Prostaglandin E, methoxime, 132-133 22 Prostaglandin F, 89-118 Methyl 2,3,5-tri-0-acety14-pentulosonate, Protoporphyrin-IX, 170,182,187 22 dimethyl ester, 187 Methyl triO-acetyl-Lthreonate, 20 Psicofuranine, 388,394 Mitomycin C, 455457 Psicose, 388 Mtorubrin, 532 D-Psicosyl chloride, 394 Monomethyl die-acetyl-L-tartrate, 20 Psoralidine, 5 15 Munduseran, 505 Pterocarpin, 522,524 Munduserone, 500,509 Puromycin, 388,390 DL-Mycarose, 62,64 Pyrroporphyrins, 174 Pynoporphyrin-XV, 173 Nafcillin, 338 Pyrroporphyrin-XVIII, 179 Neocycloheximide, 453
Reaction Index (-)Quinic acid, 70,71 (?)-Quinide, 71 Rhodin-g, ,209 Rhodoporphyrin, 169 Rhodoporphyrin-XV, 173, 181 dimethyl ester, 189,191 Ribitol, 1 1 Ribose, 6 D-Ribose, 24,64 DL-Ribose, 13-15,73 Rotenone, 503-504 Sangivamycin, 388,398 Sativol, 515 Sclerotioramine, 526 Scriblitifolic acid, 551 Serratamolide, 424426 Seselin, 470,471 Shikimic acid, 71 Siccanochromen A, 478 L-Sorbose, 64 Stercobilin, 218 Stercobilin-IIIa, 222 Stercobilin-Xllla, 219 Sterigmatocystin, 485498 Streptidine, 365 Streptose, 365 Streptozotocin, 380 Terramycin, 359 Tetracycline, 356 Tetrahydropyrano [ 2,341-1,4-dioxane,
31
Tetrahydrosclerotioramine, 527 Tetrahydrosclerotiorin, 529 1,3,4,5-Tetrahydroxy-l cyclohexanecarboxylic acid, 70 ‘letramethoxycoumestan, 515 1,2,3,7-Tetramethoxyxanthone,540 Tetrose, 6 Tetrulose, 6 DL-glycero-Tetrulose, 19 3-endo-Thiomet hyl-3-exo.cyano-2-thiobicyclo (2.2.1.1 hept-Sene, 28 Thiophanic acid, 546 L-Threitol, 20 DL-Threitol, 17
603
L-Threonic acid, 21 L-Threono-l,4-lactone, 20 D-Threose, 19 Thuringion, 546 Toyocamycin, 388,398 3,4,6-Trideoxy-3dimethylamino-D-xylohexose, 47
2,3,6-Trideoxya-DL-hex-2enopyranos4ulose, 56 2,3,6-Trideoxy-L-glycero-hexopyranose4ulose, 56 Trifoliol, 519 3,8,9-Trimethoxypterocarpin,522 Tubercidin, 388,398 Tyrocidine A,412413 Tyrothricin, 406413 Ubichromenol, 476 d-Urobilin, 225 Wrobilin, 218 Urobilin-111, 226 Urobilin-IXa, 215,226 Urobilin-XIII, 226 Uroporphyrin-I, 174
Uroporphyrin-I1,167,169,174
Uroporphyrin-Ill, 169, 174 Uroporphyrin-IV, 169,174
Valgramicidin A, 407409 4-Vinyldeuteroporphyrin-IXdimethyl ester, 189 2-Vinyldeuteroporphyrin-IXmethyl ester, 187
2-Vinylrhodoporphyrin-XV dimethyl ester,
189
Vitamin B,, ,270 Wedelolactone, 514-516 Xanthobilirubic acid, 213,214 Xanthoxyletin, 476 Xanthyletin, 470 DL-Xylose, 6,13 Yeast alanine t-RNA, gene, 321-324 Zearalenone, 427634
